Decoding Complex Texture of Starchy Foods During Oral Processing: Electromyographic Insights and Future Horizons

Abstract

Starchy foods are characterized by complex internal structures and dynamic oral transformations, including mechanical breakdown, saliva-mediated hydration and enzymatic hydrolysis. These processes play a central role in texture perception but cannot be fully simulated by conventional texture evaluation methods. Electromyography (EMG) records bioelectrical signals generated by masticatory muscles and provides an objective approach to characterizing muscle recruitment patterns, chewing load, and oral processing strategies during actual food consumption. This review summarizes recent advances in the application of EMG to starchy food research, with particular emphasis on its advantages in evaluating key textural attributes, including hardness, crispness, stickiness, and springiness. The roles of time- and frequency-domain EMG features in revealing the biomechanical responses of different starchy foods during oral processing are also discussed. In addition, this review highlights the value of EMG in monitoring chewing rhythm and swallowing behavior, as well as in exploring the relationship between oral processing and digestive health. Its potential applications in personalized nutrition guidance, healthy food development, and customized food design for individuals with dysphagia are further considered. Despite its advantages, the broader application of EMG remains constrained by challenges in multichannel signal acquisition, high-dimensional data processing, and physiological interpretation. Future integration of EMG with texture analysis, rheological measurements, sensory evaluation, and artificial intelligence may enable more comprehensive and intelligent food texture evaluation systems. Moreover, EMG-derived information on muscle recruitment and masticatory force regulation may provide physiological insights for the development of biomimetic chewing systems and robotic oral processing functions.

Graphical abstract

Keywords

EMG / Starchy foods / Food texture / Oral processing / Texture perception

Highlight

	● Starchy foods show dynamic texture evolution during mastication.
	● EMG tracks real-time muscle responses to dynamic textures of starchy foods.
	● EMG bridges starch structure, oral processing and texture perception.
	● Integrating EMG with multimodal sensing may advance embodied intelligence.

1 Introduction

Starchy foods constitute a major component of global dietary patterns and represent a primary source of dietary energy intake [1]. Their unique viscoelasticity and shear-thinning behavior give rise to a complex texture experience, which lies at the very core of their appeal to consumers. The textural complexity of starchy foods is reflected not only in the diversity of their initial oral sensations, but also in the continuous structural transitions that occur during oral processing. Biscuits, bread, rice products and glutinous rice foods, for example, may initially be perceived as crisp, crumbly, springy or viscoelastic. As mastication proceeds, however, the starch-dominant multi-component matrix progressively undergoes fracture, water uptake, saliva incorporation, softening, adhesion, aggregation and bolus formation [2,3]. Consequently, texture perception gradually shifts from the sensation associated with the first bite to the texture experienced during later stages of chewing and ultimately to swallowing suitability. These changes should not be regarded simply as general dynamic texture changes. Rather, they are closely linked to the specific physicochemical and structural characteristics of starch-rich matrices. In protein dominated foods, oral processing is often governed by the fracture, deformation or disintegration of fibrous structures, gel networks or protein aggregates [4]. In lipid dominated foods, texture perception is more strongly associated with melting, lubrication, emulsion breakdown and the release of creamy or fatty sensations [5]. In contrast, starchy foods are simultaneously influenced by mechanical breakdown, saliva-mediated hydration, salivary α-amylase activity, and the restructuring of starch granules or gelatinized starch networks within the oral cavity. Non-starch components further modulate this process by affecting matrix continuity, water binding, interfacial interactions, lubrication, and bolus formation. Their oral processing therefore represents a time-dependent evolution involving structural disintegration, saliva penetration, enzymatic softening, changes in adhesiveness and aggregation, and bolus restructuring.

This starch specific oral processing pathway indicates that the texture evaluation of starchy foods cannot rely solely on static instrumental measurements, such as single compression, shearing or puncture tests. Although conventional texture analyzers can characterize mechanical properties under defined testing conditions, they are limited in their ability to capture the continuous transformations that occur during real mastication. Sensory evaluation is inherently subjective and often requires trained panelists, making it time-consuming and difficult to standardize across individuals and laboratories. In our previous work, we developed a biphasic mechanical evaluation strategy based on stress relaxation and rheological properties to characterize the complex waxiness attribute of foods [6]. Deblais et al. further demonstrated that the perception of thickness in liquid foods follows the logarithmic relationship between stimulus intensity and perceived intensity described by the Weber-Fechner law [7]. These findings indicate that texture evaluation is highly complex and remains difficult to achieve in a rapid, real-time, and objective manner. Electromyography (EMG), as a bioelectrical signal acquisition and analysis technique, provides an objective approach for continuously monitoring the activity of masticatory and perioral muscles, such as the masseter and temporalis, throughout oral processing [8]. By capturing muscle activation patterns during chewing, EMG enables real-time assessment of chewing effort and oral processing behavior as starchy foods progress from initial biting and structural breakdown to hydration, saliva incorporation, enzymatic softening and bolus formation [9,10]. This is particularly relevant to starch-rich matrices, whose texture evolves under the combined effects of mechanical fragmentation, saliva-mediated hydration, shear-induced thinning and salivary α-amylase hydrolysis [11-13]. Therefore, EMG offers a complementary physiological basis for linking starch structure, dynamic texture perception, mouthfeel and potential consumer acceptance during real eating conditions.

EMG was introduced into food science in the 1980s as a tool for investigating oral processing mechanisms [14]. EMG typically records the electrical activity of masticatory and swallowing-related muscles, including the masseter, temporalis, and infrahyoid muscles, using bipolar surface electrodes [15]. After amplification, filtering, and digital acquisition, EMG signals can be processed to extract parameters such as the area under the EMG curve, signal amplitude, chewing duration, chewing cycles, and swallowing-related temporal variables [16]. These parameters provide information on masticatory work, muscle activation intensity, and oral processing behavior. Recent advances in EMG have improved the real-time assessment of oral muscle activity, offering new opportunities to optimize texture perception, chewing demand, and palatability of starchy foods. EMG-based evaluations have been applied to various starch-rich products, including wheat-based noodles, bread, and biscuits; rice-based cooked rice and rice cakes; and potato-based mashed potatoes and chips [17-19]. By extracting time- and frequency-domain features and characterizing the temporal patterns of muscle activation, EMG can help characterize texture-related oral processing responses associated with hardness, crispness, stickiness and springiness. When integrated with structural and physicochemical analyses, it can further help interpret the deformation, fracture, and saliva-mixing behavior of starchy foods during oral processing, thereby supporting rational formulation improvement and texture design. In addition to texture assessment, EMG has been used to investigate chewing rhythm, swallowing behavior, and eating efficiency, linking oral processing patterns to sensory preference and consumer acceptance [20]. Comparisons of chewing cycles, muscle activation intensity, and pre-swallowing oral processing time across different starchy foods can reveal the mechanical and temporal determinants of eating experience [21]. Such information is valuable for designing customized foods with enhanced palatability and swallowing safety for specific populations.

Driven by the growing demand for objective evaluation of oral texture and physiological responses in starchy foods, recent advances in EMG-based approaches are reviewed here, with a focus on their methodological principles, applications in texture evaluation, and future potential in intelligent food design. First, the fundamental principles of EMG and its implementation strategies in oral dynamics research are briefly introduced. Next, the applications of EMG in characterizing key textural attributes of starchy foods, including hardness, springiness, stickiness, and crispness, are discussed from the perspectives of muscle activation patterns, chewing load, and oral processing behavior. In addition, the use of EMG for tracking chewing patterns, evaluating oral processing efficiency, monitoring swallowing behavior, and assessing related health implications is systematically summarized. Finally, future perspectives on the integration of EMG with artificial intelligence and embodied intelligence are proposed. Therefore, this review focuses specifically on starch-dominant multi-component foods, because their oral processing involves not only the mechanical and enzymatic breakdown of starch structures, but also matrix-level regulation by non-starch components, water distribution, lubrication, and bolus formation behavior. It further highlights EMG as an objective physiological tool for texture evaluation, rational product design, healthy food development, and personalized nutrition strategies.

2 Structural Origins of Sensory Texture Complexity and Evaluation Challenges in Starchy Foods

The complex textural evolution of starchy foods during oral processing is governed by the multiscale structure of starch and its dynamic responses to the combined effects of mastication, salivary lubrication, and enzymatic hydrolysis (Fig. 1). Accurate characterization of this process is essential for understanding the mechanisms underlying texture perception in starchy foods.

During oral processing, starchy foods are exposed to shearing, compression, saliva lubrication, and salivary amylase hydrolysis, which drive hierarchical restructuring of starch from granule architecture to molecular chains and gel networks. These changes include hydration, swelling, and leaching of granules, alterations in crystalline and amorphous lamellae, molecular chain degradation, and progressive reorganization of the gel network during mastication.

2.1 Multiscale structural basis of texture complexity

During gelatinization, starch granules absorb water and swell, while crystalline regions progressively dissociate, forming a gel network with characteristic rheological properties. Acting alone or in interaction with other biopolymers such as proteins, dietary fiber, and lipids, this network is a major determinant of texture attributes [22]. The textural complexity of starchy foods fundamentally arises from the multiscale structural organization of starch. Starch is a hierarchically organized system spanning molecular, crystalline, and granular levels. The amylose/amylopectin ratio, chain length distribution, and degree of molecular entanglement regulate water absorption, swelling, gelatinization, and retrogradation, thereby shaping textural attributes such as hardness, crispness, and adhesiveness [23]. Native starch generally exhibits A-, B-, or C-type crystalline patterns. Differences in crystalline polymorphs and the degree of double-helical order govern the responses of starch granules to heat and water, thereby affecting gelatinization onset, structural disintegration, and subsequent network reorganization [24]. Granule size, morphology, surface properties, and the spatial arrangement of crystalline and amorphous regions further determine fracture behavior and fragment release under heating, shear, and mastication, thereby influencing attributes such as graininess, mealiness, and smoothness [25,26]. Long starch chains distributed across both crystalline and amorphous regions generally enhance gel stability and elasticity, whereas starch systems enriched in short branched chains tend to swell more readily, resulting in softer and stickier textures [27]. Chain length distribution of amylopectin also affects double helix stability and crystalline rearrangement, thereby altering network uniformity, resilience, and water holding capacity [28]. In addition, proteins, lipids, dietary fiber, and water can modulate matrix continuity, hydration, lubrication, adhesion, and bolus formation, thereby further shaping the macroscopic texture of starch-rich foods. The structural complexity of starch is progressively translated into the multidimensional texture observed at the macroscopic level. In real starchy foods, non-starch components regulate gel formation through specific molecular and matrix-level interactions. Lipids, especially fatty acids or emulsifiers, can form V-type complexes with amylose, thereby restricting starch granule swelling and amylose leaching, improving paste stability under heating and shear, and modifying retrogradation behavior [29]. Proteins may either weaken starch gels by competing for water and disrupting starch−starch hydrogen bonding, leading to looser networks and lower gel strength, or reinforce the matrix when they participate in continuous protein−starch networks [30]. Moreover, starch, proteins, and lipids can form ternary complexes with higher short-range order and crystallinity than binary starch−lipid complexes, further altering viscosity development, gel elasticity, and bolus-forming properties during oral processing.

2.2 Texture evolution induced by oral processing

Starchy foods exhibit pronounced dynamic complexity during oral processing. This complexity arises not only from the intrinsic multiscale structure of starch, but also from its continuous reorganization and evolution under the combined effects of mastication, salivary wetting, and α-amylase activity. During mastication, compressive and shear forces disrupt the gel network of starchy foods, leading to matrix fragmentation and greater exposure of internal domains to saliva. The resulting increase in contact area accelerates saliva infiltration, water penetration, and enzymatic hydrolysis [31]. The wettability of starch is governed by the coordinated effects of multiscale structural features. The abundant hydroxyl groups in starch molecules confer intrinsic hydrophilicity, whereas the degree of crystalline order and the proportion of amorphous regions regulate the rate of water penetration. Different crystalline polymorphs show distinct responses to water. A-type starch has a relatively compact crystalline structure, whereas B-type starch possesses a more open and hydrated crystalline arrangement, which can facilitate water access and diffusion. Rapid water penetration into amorphous regions promotes swelling and may further induce transformation of the original crystalline structure [32]. Expanded and brittle porous starchy foods provide a representative example. After wetting, the loose and porous framework of fresh popcorn exhibits ultra-fast water uptake and rapid structural collapse. Jiang et al. reported that the three-dimensional interconnected microporous network generated during popping provides continuous capillary pathways in both horizontal and vertical directions, thereby accelerating liquid spreading and infiltration [33]. In addition to this capillary effect, the authors attributed the rapid wetting behavior to the presence of water-soluble crystallized starch in fresh popcorn, which further promotes water spreading and absorption. Upon exposure to humid air, this soluble crystalline state gradually transforms into an insoluble amorphous state, leading to impaired hydrophilicity and slower wetting. As a result of the synergistic effects of the porous architecture and starch crystalline state, a water droplet can completely wet fresh popcorn within 40 ms, accompanied by immediate structural collapse. This rapid wetting-collapse process helps explain the characteristic transition from initial crispness to softening during oral processing. Such dynamic textural changes during oral processing, jointly governed by wettability, mechanical breakdown, and enzymatic hydrolysis, cannot be adequately captured by conventional methods based solely on static structural evaluation.

2.3 Main methods for texture evaluation

The major approaches for evaluating the textural quality of starchy foods include instrumental texture analysis, rheological measurements, and sensory evaluation. Texture analyzers characterize the macroscopic mechanical responses of foods under external forces. By applying compression, puncture, shear, tensile, or texture profile analysis (TPA) tests, force-time or force-displacement curves can be recorded to derive parameters such as hardness, cohesiveness, springiness, adhesiveness, chewiness, and gumminess [34]. These methods are suitable for characterizing the overall textural properties of solid and semi-solid foods (Fig. 2A). Rheometers, in contrast, focus on material flow and deformation under continuous shear or oscillatory loading. By applying shear force or oscillatory stress or strain, they measure parameters such as apparent viscosity, yield stress, storage modulus (G′), and loss modulus (G′′) [26], and are therefore more suitable for characterizing the viscoelasticity and structural stability of semisolid, colloidal, and other complex food systems (Fig. 2B). In essence, both methods are ex vivo instrumental measurements that reflect the physicochemical properties of samples under predefined mechanical conditions. Sensory evaluation provides direct information on human perception, but it is inherently subjective, requires trained panelists, and may suffer from limited reproducibility across individuals and laboratories [2]. By contrast, EMG assesses texture-related physiological responses from the perspective of human oral processing. By recording the electrical activity of masticatory muscles, such as the masseter and temporalis, during chewing, EMG enables the extraction of parameters including chewing cycles, chewing duration, signal amplitude, and integrated EMG activity [15]. Compared with texture analyzers and rheometers, the major advantage of EMG lies in its ability to bridge food physical properties with human oral physiological responses. This advantage is particularly relevant for starchy foods, which undergo dynamic textural evolution in the oral cavity, including brittle fracture, saliva-induced wetting, structural softening, and enzymatic hydrolysis. During these processes, EMG can capture real-time adjustments in masticatory muscle activation patterns, thereby linking the temporal evolution of food texture with physiological responses during oral processing and helping to reveal the physiological mechanisms underlying texture perception (Fig. 2C). Therefore, by capturing individual physiological responses to food texture during real oral processing, EMG helps address the limitations of conventional methods in monitoring dynamic oral processing changes and inter-individual differences in texture perception. It can also work synergistically with texture analysis, rheological measurements, and sensory evaluation to establish a more comprehensive framework for food texture assessment.

3 Methodology of EMG: Signal Acquisition, Processing and Analysis

Electromyographic signals reflect the summed action potentials of skeletal muscle fibers activated by spinal α-motor neurons under central motor control [35]. They can be non-invasively detected by surface electrodes during functional movements such as mastication and swallowing [36]. A complete EMG system typically comprises electrodes, amplifiers, analog-to-digital converters, and signal processing units [37]. The electrodes, which serve as the signal acquisition interface, are typically disposable silver chloride (Ag/AgCl) surface electrodes. In oral dynamics research, as shown in Fig. 3A, electrodes must be precisely positioned on the muscle belly of the target muscles (e.g., masseter, anterior temporalis, anterior belly of the digastric muscle, and genioglossus muscle) based on strict anatomical standards, with appropriate inter-electrode spacing (usually 5−20 mm) to optimize signal quality and minimize cross-talk from adjacent muscles [38,39]. Given the weak nature of the raw EMG signals collected from the skin surface (typically in the microvolt range), which are susceptible to environmental noise interference, high-performance bioelectrical amplifiers are employed to initially amplify and pre-filter the signals, suppressing baseline drift and high-frequency noise. Subsequently, an analog-to-digital converter (ADC) is used to convert the analog signals into digital form, where the sampling rate (typically no less than 1000 Hz) and resolution (e.g., 16-bit or 24-bit) are critical parameters determining signal detail and dynamic range [5,40].

After amplification and digitization, EMG signals are commonly processed by band-pass filtering, typically within 20−500 Hz, to attenuate motion artifacts and high-frequency noise, together with notch filtering at 50 or 60 Hz to reduce power-line interference, followed by full-wave rectification and smoothing, such as root mean square calculation (Fig. 3B). Commonly used software packages for EMG signal acquisition and processing include OpenSignals and LabChart 8 [41,42]. Time-domain analysis of the signals from different muscles allows for precise depiction of the activity patterns of various muscle groups during chewing and swallowing (Figs. 3C-3D). EMG-based oral processing parameters include time-domain metrics, such as integrated EMG, which reflects cumulative muscle activity; frequency-domain metrics, such as spectral energy distribution, which characterizes muscle activation patterns and fatigue-related changes; and temporal metrics, such as the number of chewing cycles, chewing cycle duration, and muscle activation latency, which reveal neuromuscular coordination patterns [43-45] (Fig. 4 and Table 1). A reference or ground electrode is commonly placed on the wrist or back of the hand to reduce electrical noise, whereas inter-individual differences in muscle activity are mainly addressed through appropriate normalization procedures, such as normalization to maximum voluntary contraction (MVC) or to a reference sample [46]. To account for individual preferences or habitual differences in chewing, it is essential to apply a standardization procedure using an appropriate reference material before the experiment begins. Furthermore, experimental data should be normalized to ensure comparability across individuals [47]. For instance, EMG parameters under different food conditions can be converted into percentages based on a reference sample [48]. This normalization is considered more clinically relevant than using raw surface EMG data alone.

4 Textural Properties and Sensory Experience Assessed by EMG

EMG can link the textural attributes of starchy foods to oral processing responses through parameters such as amplitude, integrated EMG, burst duration, chewing cycles, and total mastication time. During mastication, starchy foods undergo continuous structural and physicochemical transformations. Therefore, changes in EMG signals reflect the neuromuscular demand imposed by the combined effects of multiple evolving texture attributes, rather than the variation of a single property. The initial hardness or crispness of starchy foods may mainly determine the early chewing load and EMG amplitude, whereas saliva incorporation and starch hydration may progressively reduce fracture resistance while increasing adhesiveness, cohesiveness, or elasticity in the middle and late stages of chewing. Based on this dynamic evolution, hardness, crispness, stickiness, and springiness are selected as representative texture dimensions in this review to clarify the main contribution of different texture attributes to EMG responses and to examine the dynamic oral processing trajectories of starchy foods and the associated neuromuscular regulation strategies. Detailed literature-reported EMG parameters associated with textural attributes in different starchy food products are provided in Table 2, whereas the main relationships among texture categories, oral processing stages, and EMG signal traits are summarized in Table 3. Fig. 5 further illustrates the dynamic evolution of representative texture attributes and their corresponding neuromuscular responses during oral processing.

4.1 Attributes related to hardness

Hardness is one of the key sensory attributes influencing consumer acceptance of solid foods and is typically defined as the force required to deform or fracture a food sample [49]. Traditional instrumental measurements, such as TPA, characterize hardness by recording the maximum peak force when the probe first compresses the sample [27]. As food hardness increases, greater masticatory muscle activation is required to overcome the increased resistance. Mechanoreceptors in the periodontal ligament detect this resistance and provide feedback to the central nervous system, promoting the recruitment of additional motor units. Consequently, EMG signals associated with harder foods typically exhibit higher amplitudes and broader peaks [50]. The hardness of different matrices exhibits distinct time-frequency characteristics in EMG signals (Fig. 5A). In a study by Rustagi et al., involving various textured foods including dhokla, principal component analysis (PCA) revealed that the PC1 was predominantly composed of total muscle activity, chewing duration, and cycle duration, explaining up to 82% of the data variance and showing a significant positive correlation with both sensory and instrumentally measured hardness [51]. This suggests that harder foods require greater cumulative masticatory muscle activity, as reflected by higher integrated EMG (iEMG) values. In biscuit texture evaluation, mastication time was also identified as an effective predictor of perceived hardness. Harder biscuits required longer chewing sequences to provide sufficient mechanical work for reducing the food matrix to a swallowable bolus [52].

Hardness also affects masticatory cycle-related parameters. A study by Kaur et al. on different rice formulations showed that increased rice hardness not only increased the number of chews but also altered the duration of individual chewing cycles [53]. Kohyama et al. further showed that high-amylose rice induced longer burst duration and higher EMG amplitude during the early and middle stages of chewing, which was attributed to its greater hardness [54]. Unlike the rapid force drop associated with brittle fracture, hardness-dominated chewing is characterized by sustained and progressive structural rupture, leading to slower attenuation of muscle activity throughout the chewing sequence. Overall, by capturing both the amplitude and duration of muscle contractions, EMG can objectively characterize hardness-related attributes of starchy foods and the associated oral processing load.

4.2 Attributes related to crispness

In low-moisture starchy foods, crispness arises from rapid brittle fracture of a rigid glassy matrix formed during drying, baking, or frying [55]. During chewing, this matrix undergoes successive brittle fractures, generating characteristic mechanical and acoustic responses. Traditional instrumental tests, such as three-point bending and puncture tests, can simulate the initial bite-induced fracture and measure fracture force and acoustic signals [56]. However, these conventional instrumental methods are limited in their ability to fully replicate the dynamic oral environment during human chewing, particularly the real-time effects of saliva infiltration and complex jaw movements on crispness perception [57]. During the chewing of crispy foods, rapid brittle fracture requires brief and coordinated force application to initiate structural breakdown, which is reflected in EMG signals as short burst duration and sharp activation patterns. Different starchy foods elicit distinct chewing strategies (Fig. 5B). For hard and crisp starchy foods, such as biscuits, combined EMG and kinematic analysis showed that the chewing pattern primarily shifted from vertical compression in the early stage to horizontal shearing in the later stage [58]. Compared with crisp fruits and vegetables, which are primarily processed through vertical compression, biscuits require greater biting force to initiate fracture before salivary softening occurs. Further studies by Sodhi et al. on different biscuit formulations showed that mastication time was significantly positively correlated with sensory fracturability and hardness [52]. Specifically, biscuits with denser structures and higher perceived fracturability tended to require longer chewing sequences.

By comparison, in thin or puffed crispy starchy foods, such as potato chips and corn chips, crispness is more closely associated with microstructural porosity. High-porosity starch matrices can generate intense acoustic signals upon slight deformation, even at relatively low hardness [56]. The fine structure of amylose, especially the porous and glassy rigid network formed during frying, plays a key role in supporting and promoting brittle fracture in potato chips [59]. These characteristics are reflected in oral processing as reduced chewing effort and shorter chewing sequences. In brittle cereal foods, Hedjazi et al. showed that mastication was dominated by compression rather than shearing, and that products requiring fewer chewing cycles and shorter sequence durations tended to have higher sugar content [60]. This was attributed to the formation of a more brittle structure or finer cellular architecture, which may facilitate faster fragmentation and bolus formation. A study by Luckett et al. on potato chips with varying levels of crispness found that sensory crispness was negatively correlated with average chew work but positively correlated with the number of chews [61]. Chips with higher crispness may fracture more readily, thereby reducing the muscular load required for each chew; however, the resulting dry and fragmented particles may require additional chewing and saliva incorporation before forming a swallowable bolus. In summary, EMG can capture dynamic changes in masticatory muscle responses during the chewing of crispy starchy foods, from initial brittle fracture to later saliva-induced softening and bolus formation, thereby providing physiological insights into the mechanisms underlying crispness perception.

4.3 Attributes related to stickiness

Stickiness in starchy foods arises from cohesive forces within the food bolus and adhesive interactions between the food matrix and oral surfaces, such as the teeth and palate [62]. The starch gel network, stabilized by hydrogen bonding and other weak interactions, enhances both cohesion within the bolus and adhesion to oral surfaces. Amylopectin-rich systems tend to exhibit greater stickiness, partly because their highly branched molecular architecture promotes water retention and intermolecular entanglement [63]. Unlike hardness or crispness, which elicit strong EMG responses during the initial stages of chewing, stickiness affects EMG signals in a delayed and persistent manner (Fig. 5C). EMG studies indicate that the influence of stickiness on chewing behavior becomes most evident during the middle to late stages of the chewing sequence [64]. For sticky starchy foods, such as rice cakes and cooked rice, the food matrix gradually hydrates and forms a cohesive and adhesive bolus during chewing, requiring sustained masticatory muscle effort to overcome adhesive resistance. In EMG signals, this resistance is typically manifested not as a single burst peak, but as prolonged muscle activity, reflected by increased chewing cycles and total chewing duration. Iguchi et al. compared the chewing characteristics of cooked rice and rice cakes, showing that the greater cohesiveness and adhesiveness of rice cakes resulted in more chewing cycles and a longer total chewing time before swallowing than cooked rice [65]. This difference was mainly attributed to the rheological properties of the foods. During chewing, the activity of the masseter and mylohyoid muscles changed progressively and was influenced by textural factors such as hardness, stickiness, and cohesiveness.

Furthermore, stickiness modifies mastication effort by altering the pattern and duration of muscle activity during oral processing. Kohyama et al. reported that, in cooked rice samples with different water contents and milling degrees, total mastication effort, represented by iEMG, was closely associated with the texture balance of rice, defined as the ratio of stickiness to hardness [66]. Rice with greater stickiness required prolonged muscle activity until the bolus reached a swallowable state. It is worth noting that sticky foods may also induce specific activation of jaw-opening muscles, such as the digastric muscle, as well as tongue muscles. Highly sticky starchy foods tend to adhere to the teeth and oral surfaces, requiring additional muscle effort during the jaw-opening phase to separate the dental arches. During the pre-swallowing clearance phase, sustained tongue muscle activity may also be required to remove starch residues from the oral mucosa. Kohyama et al. reported that excessive stickiness in rice cakes increased swallowing difficulty, as indicated by changes in the amplitude and duration of suprahyoid EMG activity during swallowing [67]. These findings suggest that EMG can characterize stickiness-related oral processing responses beyond mastication, including prolonged muscle activity, bolus clearance, and swallowing-related effort. Thus, EMG provides a physiological basis for evaluating the processing load and perceptual consequences of stickiness in starchy foods.

4.4 Attributes related to springiness

Springiness refers to the ability of a food to recover its original shape after the removal of an external force and is commonly observed in gel-like and dough-based starch foods, such as rice cakes (Niangao), noodles, and fermented bread. In these systems, gelatinized starch molecules, together with other matrix components, form a resilient network through hydrogen bonding, molecular entanglement, and physical association [68]. In springy foods, the structure does not collapse abruptly during chewing; instead, it undergoes continuous deformation under tooth compression, followed by partial recovery or rebound during unloading [69]. This physical property is associated with a distinct neuromuscular control strategy during chewing, which is reflected in EMG signals as specific temporal patterns and muscle activation profiles (Fig. 5D). Springiness markedly affects per-cycle masticatory effort. In a comparison between cooked rice, which is hard and granular, and rice cakes, which are softer but highly viscoelastic, cohesive, and adhesive, rice cakes showed lower instrumental hardness than cooked rice but induced significantly higher masseter iEMG per chewing cycle [18]. This may be because, during the chewing of highly springy starch matrices, the jaw-closing muscles must not only overcome the yield resistance of the food but also continuously counteract the rebound and deformation resistance generated by the starch network. Unlike brittle foods, which can fracture rapidly during compression, springy foods require sustained muscle tension during the closing phase, resulting in greater per-cycle masticatory effort.

Foods with pronounced springiness can prolong the chewing sequence and delay the attenuation of EMG activity. Using EMG in combination with jaw-movement kinematics, Kohyama et al. performed cluster analysis on foods with different textures and showed that springy foods, represented by weak gels such as konjac gel, exhibited chewing characteristics distinct from those of hard and brittle foods [64]. Foods with pronounced springiness typically require more chewing cycles to reach the swallowing threshold. During the middle to late stages of chewing, EMG amplitude and burst duration tend to decrease more gradually, probably because the elastic structure requires repeated grinding and compression to disrupt the internal gel network. Thus, burst duration may serve as a sensitive indicator for distinguishing springiness-related responses from hardness-dominated chewing behavior. In a study on different formulations of idli, a fermented sponge-like rice cake, Dhillon et al. reported that middle burst duration in the EMG signal was strongly correlated with sensory cohesiveness and springiness [70]. For starchy foods with a sponge-like porous structure, the matrix undergoes gradual densification under tooth compression rather than abrupt fracture or penetration. This results in broader EMG burst waveforms than those observed for brittle foods, together with longer tooth contact time. By capturing sustained muscle activity against rebound resistance, reflected by higher per-cycle iEMG, and delayed structural breakdown, reflected by longer chewing sequences and broader waveforms, EMG provides a quantitative approach for characterizing springiness-related attributes in starchy foods.

5 EMG Monitoring of Oral Processing Dynamics

EMG technology not only characterizes the textural attributes of foods but also enables real-time and continuous recording of muscle activity, thereby revealing the dynamic processes of oral food processing. This section focuses on the application of EMG in monitoring the formation and breakdown of starchy food boluses, quantifying subjective chewing sensations and preferences, evaluating the effects of mastication on digestive health, and assessing swallowing safety and efficiency in specific populations, such as individuals with dysphagia.

5.1 Bolus formation and breakdown during oral processing

Starchy foods undergo dynamic textural transformation during mastication, driven by the combined effects of saliva and mechanical forces. Soft gel foods experience localized network breakdown and structural reorganization, leading to gradual bolus homogenization, whereas crispy foods initially undergo brittle fracture and subsequently shift toward softened deformation as rapid water uptake promotes plasticization [71,72]. To analyze this complex dynamic process, studies typically divide the chewing sequence into early, middle, and late stages to characterize the evolution of bolus structure and oral processing behavior [73-75]. In the early stage of mastication, the amplitude and frequency characteristics of EMG signals mainly reflect the initial textural properties of food and its instantaneous deformation behavior. A study by Kaur et al. showed that parboiled rice, due to increased hardness caused by starch retrogradation, significantly elevated the amplitude and burst duration of masseter muscle activity during the early chewing phase, whereas cooked rice, with its higher degree of gelatinization, reduced the mechanical work required for structural breakdown [76]. As chewing progresses into the middle and late stages, the gradual breakdown of the starch matrix microstructure by salivary enzymes and chewing forces is typically reflected in a decrease in EMG amplitude and a more regular chewing rhythm. However, this dynamic decay pattern is highly dependent on the rheological properties of starch. A comparative study by Shiozawa et al. found that the EMG activity of regular rice significantly decreased during the late chewing stage, indicating that the food bolus had been broken down and softened. In contrast, rice cakes with high viscoelasticity maintained consistent EMG activity throughout the entire chewing cycle, with no significant attenuation [18]. This reveals that EMG can sensitively monitor whether the gel network undergoes structural breakdown or simply undergoes deformation while maintaining its cohesion. Electromyographic characteristics during the swallowing preparation phase, such as rhythmic activation of the mylohyoid muscle, signal that the bolus has reached the appropriate viscoelastic state for swallowing, marking the end of oral processing. Through clustering analysis and principal component analysis of these dynamic parameters, EMG technology not only quantifies the texture of starch foods but also reproduces the adaptive chewing strategies of individuals when faced with foods of varying gel strength [77].

The integration of multiple technologies, such as simultaneous recording of jaw movement trajectories, tongue pressure [64], and real-time video fluoroscopic swallowing studies [78], enables a more comprehensive analysis of the correlation between chewing muscle activity and bolus deformation. Kohyama et al. systematically demonstrated that while EMG technology can assess overall chewing force and work, combining it with jaw kinematic monitoring enables the differentiation and quantification of three independent dimensions of chewing control: rhythm control related to the chewing cycle (PC1), jaw movement amplitude modulated by bolus mechanical resistance (PC2), and chewing force reflecting muscle strength (PC3) [64]. This finding demonstrates that synchronized monitoring technology not only identifies "whether the chewing process is effortful" but also clarifies "how it is effortful"—that is, how the brain dynamically adjusts the spatial trajectory, timing, and force of jaw movements as a coordinated strategy to precisely process foods with different textures. This dynamic perspective aids in understanding the unique behavioral pathways of different starch matrices during oral processing, providing a theoretical foundation for designing foods with specific textures and swallowing properties.

5.2 Prediction of texture acceptability and sensory preference

Consumer acceptability and overall preference are rarely determined by static texture metrics alone; instead, they are shaped by the dynamic evolution of bolus structure and lubrication during oral processing [79]. EMG is particularly valuable because it translates these dynamics into quantifiable features. On the one hand, the amplitude or peak level of masticatory muscle activity relates to instantaneous maximal bite demand, serving as a physiological proxy for mechanical load [80]. On the other hand, muscle activity per chew (mV·s) and total muscle activity (mV·s) more closely represent chewing work or overall effort and have been reported to correspond closely to instrumentally measured work of fracture [81,82]. These metrics therefore help explain how negative texture experiences, such as hard to chew or effortful, reduce acceptability and preference. The relationship linking texture to oral burden and subsequent preference also holds in realistic starchy staple systems. Using cooked rice from conventional japonica rice and japonica rice with high resistant starch content as models, Zhao et al. compared rice texture, chewing behavior, EMG features, and overall preference, showing that rice with higher hardness and lower stickiness increased the number of chews, chewing duration, and total masticatory muscle activity, thereby increasing oral processing burden and reducing consumer liking [19]. Notably, even among samples with comparable instrumental hardness, differences in stickiness could still be clearly discriminated by EMG activity and chewing behavior, underscoring that texture acceptability is highly sensitive to dynamic oral processing features.

EMG helps move preference research beyond a static outcome toward an interpretable and predictable mechanism. In a representative starchy model system, mashed potato gelled with κ-carrageenan was used as a test food to compare habitual chewing with paced chewing at a fixed frequency. The fixed-frequency condition did not significantly increase the number of chews, but it prolonged chewing duration and redistributed masticatory effort across muscle groups (e.g., with a greater contribution from the temporalis), leading to systematic shifts in perceived oral texture dimensions such as moistness, adhesiveness, and cohesiveness. Interestingly, overall preference and acceptability were higher under the fixed-frequency condition, providing direct evidence that chewing mode itself can alter the acceptability and hedonic experience of starchy foods [83]. These findings suggest that, in starchy product development, EMG can serve as a bridge linking formulation, oral processing, and sensory preference by enabling objective quantification of oral processing burden using indicators such as total masticatory muscle activity, chewing work, and number of chews or chewing time per bite [84]. Recent advances in multimodal fusion strategies from affective computing suggest that physiological measurements or facial observations alone are often constrained by inter-individual variability and environmental noise [85]. By combining EMG, which reflects internal neuromuscular responses, with real-time facial expression analysis, which captures externally expressed affect, their complementarity can be exploited to substantially improve preference decoding [86,87]. Such integration can therefore yield more robust models for overall preference prediction while simultaneously capturing both the physical effort component (effortful vs. easy to eat) and the affective component of acceptability.

5.3 Role of chewing behavior in digestion and metabolic health

For starchy foods, postprandial glycemic response and digestion kinetics are not determined solely by starch composition or processing conditions. Mastication substantially modifies the bolus before it enters the gastrointestinal tract by altering particle size distribution, specific surface area, and the extent of saliva incorporation, thereby modulating starch accessibility to digestive enzymes [88,89]. Ranawana et al. quantified chew count and chewing duration using EMG and analyzed the particle-size distribution of chewed boluses [90]. They found that, in cooked rice, the proportion of fine particles was significantly associated with the 45 min glycemic peak and incremental area under the curve (iAUC). This indicates that, in particle size-sensitive starch systems, inter-individual differences in mastication may influence postprandial glycemic responses by remodeling bolus structure before gastrointestinal digestion.

In addition to mastication itself, eating style can also influence metabolic outcomes. Sun et al. compared oral processing characteristics and glycemic responses when white rice was consumed with chopsticks or a spoon, and reported a lower glycemic index when chopsticks were used [glycemic index (GI) ≈ 68] than when a spoon was used (GI ≈ 81) [91]. EMG and behavioral analyses indicated that this difference was largely attributable to smaller bite size, a greater number of bites, and a slower overall eating rate. Moreover, bite count was negatively associated with iAUC, whereas bite size was positively associated with iAUC. These findings suggest that, even when food formulation remains unchanged, modifying eating style and masticatory behavior can alter the pathway of bolus formation, thereby affecting starch accessibility to digestive enzymes and the resulting postprandial glycemic response. The balance among oral processing burden, starch digestibility, and sensory acceptability is particularly relevant for starchy foods rich in resistant starch or characterized by reduced stickiness. Zhao et al. found that high-resistant-starch japonica rice, characterized by greater hardness and lower stickiness, required more chews, longer chewing duration, and greater total masticatory muscle activity, accompanied by higher mean absolute value (MAV) and shorter burst duration [19]. Although these samples exhibited slower glucose-release kinetics during simulated gastrointestinal digestion, their overall liking was markedly reduced. These findings suggest that increasing oral processing burden may slow starch digestion, but this potential benefit may be accompanied by reduced texture acceptability. Therefore, EMG should not be regarded as a direct predictor of glycemic indices, but rather as a physiological tool for quantifying chewing behavior through parameters such as MAV, burst duration, and total muscle activity.

5.4 Swallowing burden and dysphagia assessment

Swallowing-related oral processing problems should be interpreted by distinguishing age-related physiological decline from dysphagia-related functional impairment [92]. Age-related decline mainly involves reduced oral muscle reserve, lower chewing efficiency, and weaker adaptation to texture changes, whereas pathological dysphagia is more directly associated with impaired bolus transport, abnormal neuromuscular coordination, and reduced swallowing safety. In older adults, age-related decline does not necessarily lead to overt swallowing failure, but it may increase the effort required to form a safe-to-swallow bolus. EMG studies using real foods have shown that older adults had lower muscle activity per chew than younger adults, but compensated by increasing chew number and chewing duration, resulting in comparable total pre-swallow EMG activity [93]. Stage-resolved EMG analysis further showed that EMG amplitude decreased during mastication in young adults, whereas this adjustment was less evident in older adults, suggesting reduced adaptation to progressive in-mouth texture changes [94]. In cooked rice with different hardness levels, older adults also showed higher normalized masseter activity than younger adults, indicating that they used a larger proportion of their maximal masticatory capacity even for relatively soft foods [95]. Thus, in age-related physiological decline, EMG is useful for detecting reduced chewing efficiency, increased relative muscle demand, and compensatory oral processing behavior.

Pathological dysphagia, by contrast, involves swallowing-related functional impairment and safety risks. Its mechanisms may include reduced tongue pressure, weakened swallowing-related muscles, impaired pharyngeal contraction, abnormal hyoid or laryngeal movement, and impaired coordination of bolus transport; in sarcopenic dysphagia, decreased swallowing-related muscle mass and strength may further reduce tongue strength, weaken pharyngeal contraction, and impair swallowing endurance [96]. Therefore, EMG interpretation in dysphagia should focus not only on chewing effort, but also on the activation pattern of swallowing-related muscles, particularly the submental and suprahyoid muscle groups. Longer activation duration or altered amplitude of these muscles may indicate compensatory effort or abnormal neuromuscular recruitment during swallowing [96,97]. Since texture modification, including thickening, gelation, and pureeing, is commonly used to reduce aspiration or choking risk but may reduce palatability and nutrient density, dysphagia-friendly foods should be evaluated not only by rheological properties or international dysphagia diet standardisation initiative (IDDSI) levels, but also by their neuromuscular demand during oral processing and swallowing [98-100]. For IDDSI levels 0−2 texture-modified liquid diets, increased viscosity and IDDSI level, together with lower perceived ease of swallowing, were associated with higher oral processing and swallowing-related EMG activity [101]. These findings suggest that starchy dysphagia-friendly foods, such as starch-thickened beverages and starch-polysaccharide gels, should be designed by balancing swallowing safety, sensory acceptability, and neuromuscular burden.

6 Factors Influencing EMG Responses to Starchy Foods During Oral Processing

Oral processing represents the first step of food digestion and a dynamic process involving sensory perception and central nervous system feedback. Chewing and swallowing behaviors are shaped by food texture, individual oral physiology, and eating context. In starchy foods, EMG responses during oral processing are jointly modulated by food matrix physicochemical properties, inter-individual physiological differences, processing conditions, and behavioral control strategies (Fig. 6).

6.1 Structural and physicochemical properties of the food matrix

The structural and rheological characteristics of the food matrix before and after chewing strongly shape the pattern of masticatory muscle activity. These macroscopic properties are largely governed by the molecular architecture of starch and its structural transitions during thermal processing. The amylose/amylopectin ratio is a key intrinsic factor regulating chewing kinetics. Amylose-rich systems tend to promote molecular reassociation and the formation of compact gel networks, thereby increasing hardness and shear resistance. In contrast, amylopectin contributes more strongly to viscosity, extensibility, and adhesive properties [102,103]. Starchy foods with low amylose content, such as glutinous rice, typically exhibit higher adhesiveness, which can increase the activity ratio of jaw-opening to jaw-closing muscles. This suggests that greater muscle effort is required to overcome interfacial adhesion between the food bolus and tooth surfaces, rather than being used solely for mechanical fragmentation. By contrast, high-amylose foods generally induce higher masseter burst amplitude because their firmer and more compact matrices require greater force for structural breakdown [104,105].

The molecular weight distribution of starch and the density of intermolecular networks are also important factors influencing chewing behavior. Pentikäinen et al. found that whole-grain rye bread required greater total chewing work than wheat bread [106]. This was attributed to the dense complexes formed between arabinoxylans and starch in rye, in which higher-molecular-weight polymer chains increased matrix cohesion and resistance to breakdown. As a result, participants required more frequent chewing to achieve physical disintegration of the food bolus. Starch gelatinization and retrogradation strongly influence the microscopic mechanical properties of gel networks. Starch gels with lower moisture content or those undergoing retrogradation generally exhibit increased hardness due to starch chain reassociation and recrystallization, leading to longer chewing sequences and higher iEMG values [107]. The macroscopic form of the food, such as particulate versus gel-like structures, also modulates neuromuscular control strategies. Shiozawa et al. compared particulate starch foods (e.g., cooked rice) with gel-like starch foods (e.g., rice cakes) and found that EMG activity rapidly decreased as particulate structures were broken down. In contrast, homogeneous gel foods maintained higher suprahyoid muscle activity during the later stages of chewing, likely due to their high elasticity and sustained intermolecular interactions [18]. Manda et al. further confirmed using neck EMG that starchy solids with higher hardness and elastic modulus markedly activated the posterior tongue muscles, facilitating bolus transport and formation within the oral cavity [108].

6.2 Physiological differences and individual adaptations in chewing behavior

Even when the same standardized starchy food is consumed, considerable inter-individual variability in chewing strategies can be observed. Brown et al. identified distinct chewing behavior subgroups, such as fast eaters, slow eaters, and energetic eaters, and showed that these chewing patterns were associated with different perceptions of firmness and rubberiness in model gels [109]. Salivary secretion and the oral environment are also important physiological factors contributing to variability in chewing behavior. Using generalized additive models (GAMs), Zhao et al. found that salivary flow rate and mucin concentration were non-linearly and negatively associated with chewing frequency and duration [19]. In individuals with lower salivary secretion, reduced oral lubrication efficiency may require prolonged chewing to promote hydration and agglomeration of the starchy matrix. In addition, specific physiological conditions can alter normal masticatory rhythms [110]. For example, when individuals shift from nasal to oral breathing, chewing movements may be interrupted to maintain airway patency, leading to reduced average masseter EMG amplitude and lower chewing efficiency. As a compensatory response, the number of chewing cycles may increase to achieve sufficient food breakdown.

6.3 Volitional control and external factors

Chewing is not a purely automatic reflex; it is also influenced by volitional control and external instructions, which can modulate EMG patterns. Khramova et al. demonstrated that subjects could consciously regulate their chewing frequency, thereby altering texture perception during oral processing [83]. Similarly, Okubo et al. found that, when instructed to chew well, participants did not simply increase bite force. Instead, they shortened the burst duration of each chewing cycle and reduced the load on the suprahyoid muscles, enabling finer comminution of the food bolus [41]. These adaptive behavioral adjustments highlight the importance of standardized instructions in sensory and EMG evaluations to ensure consistent and comparable data. External eating conditions and intake patterns also modulate the dynamics of chewing. Factors such as eating utensils, food size, and food form influence bite volume and the initial structure of the food bolus, thereby affecting the intensity and timing of muscle activity [48,82,91,111]. In addition, specific eating behaviors, such as noodle slurping, involve coordinated movements of the lips, tongue, cheeks, and jaw rather than simple chewing alone. These behaviors may generate EMG patterns that differ from conventional mastication, with changes in muscle activation timing, cycle duration, and rhythm.

7 Challenges and Future Perspectives

EMG is a physiological response-based approach for revealing the dynamic relationships among food structural changes, oral processing behavior, inter-individual physiological variability and texture perception. Although EMG provides an important physiological perspective for the dynamic evaluation of starchy foods during oral processing, several critical challenges remain. (1) EMG does not measure food texture itself, but records the neuromuscular response elicited by texture changes. Although EMG amplitude appears to be quantitative, it should essentially be treated as a relative measure and cannot be directly compared across individuals, or even across different experimental settings within the same individual. (2) EMG signals are influenced by both food structural properties and individual oral physiological characteristics. Chewing habits, dentition status, bite force, salivary secretion, tongue motor function and swallowing ability can all alter muscle activation patterns, meaning that the same food may induce different EMG responses across individuals. (3) EMG alone is insufficient to fully explain the dynamic texture transformation of starchy foods in the oral cavity. During mastication, these foods undergo simultaneous mechanical breakdown, saliva-mediated wetting, enzymatic hydrolysis, adhesion, aggregation, bolus formation and swallowing preparation, making it difficult to assign EMG responses to a single isolated texture attribute. Therefore, the current bottleneck in EMG research is not signal acquisition itself, but how to normalize, attribute and interpret muscle electrical activity in a mechanistically meaningful way within the context of complex food structural changes and individual physiological variability.

Future research on EMG in starchy food studies should move beyond single signal analysis toward standardized, multimodal and mechanism-guided evaluation. Standardized protocols for EMG acquisition and analysis in food oral processing research are needed, including electrode placement, muscle selection, sampling frequency, filtering procedures, normalization methods, feature extraction and reporting criteria, in order to reduce the influence of experimental settings on data interpretation. In parallel, EMG should be integrated with texture analysis, rheological measurements, acoustic signals, jaw kinematics, tongue pressure measurements, salivary characteristics, bolus structure analysis, sensory evaluation, brain electrophysiological signals such as electroencephalography, and facial expression analysis to build multimodal databases capable of describing food structure, oral processing behavior, physiological responses, and perceptual and affective responses simultaneously (Fig. 7). Although artificial intelligence and machine learning can help extract latent patterns from complex EMG signals, models may remain at the level of statistical correlation and fail to generate interpretable scientific conclusions if they are not constrained by food structure, oral physiology and sensory mechanisms. Therefore, future integration of EMG with intelligent algorithms should not focus solely on predictive accuracy, but should also address whether the model can explain why specific starch structures, degrees of gelatinization, retrogradation states and bolus formation behaviors elicit particular muscle responses. On this basis, flexible electronics, wireless sensing and wearable devices may support the transition of EMG from a laboratory analytical tool to a real-time oral processing monitoring technology, while also providing a more reliable physiological basis for biomimetic chewing systems, robotic oral processing platforms and personalized food design. By integrating EMG features, chewing trajectories, bite force, bolus formation behavior, food structural parameters, electroencephalography (EEG)-based neural responses and facial expression derived perceptual cues, future studies may establish an intelligent food texture evaluation system centered on human physiological and perceptual responses, thereby promoting the development of starchy foods toward precision, personalization and mechanism-guided design.

8 Conclusions

Starchy food texture is not a static material property, but a dynamic perceptual outcome generated through the interaction between food structure, oral processing behavior, and human physiological responses. During mastication and swallowing, mechanical breakdown, salivary wetting, enzymatic hydrolysis, bolus formation, and neuromuscular regulation jointly shape the perceived hardness, crispness, stickiness, and springiness of starchy foods. EMG provides a physiological means to capture these dynamic responses by quantifying muscle activation patterns, chewing effort, temporal coordination, and swallowing-related activity. Therefore, EMG complements conventional texture analysis, rheology, and sensory evaluation by linking food structural properties with real-time oral processing behavior. Current evidence indicates that EMG is particularly valuable for characterizing texture-related processing load, inter-individual differences, digestive implications, and swallowing difficulties in vulnerable populations. However, its broader application requires standardized experimental protocols, improved signal interpretation, and careful consideration of individual variability. With further advances in wearable sensing, artificial intelligence, and biomimetic oral processing systems, EMG may support a transition from static texture measurement toward a more physiological, individualized, and intelligent framework for starchy food evaluation and design.

References

[1] FAO. Dimensions of Need: An Atlas of Food and Agriculture. Rome: Food and Agriculture Organization of the United Nations, 1995

[2] Li H Y, Prakash S, Nicholson T M. et al. Instrumental measurement of cooked rice texture by dynamic rheological testing and its relation to the fine structure of rice starch. Carbohydrate Polymers, 2016, 146: 253–263

[3] Wang S J, Yu J L, Zhu Q H. et al. Granular structure and allomorph position in C-type Chinese yam starch granule revealed by SEM, ¹³C CP/MAS NMR and XRD. Food Hydrocolloids, 2009, 23(2): 426–433

[4] Çakır E, Daubert C R, Drake M A. et al. The effect of microstructure on the sensory perception and textural characteristics of whey protein/κ-carrageenan mixed gels. Food Hydrocolloids, 2012, 26(1): 33–43

[5] Çakır E, Vinyard C J, Essick G. et al. Interrelations among physical characteristics, sensory perception and oral processing of protein-based soft-solid structures. Food Hydrocolloids, 2012, 29(1): 234–245

[6] Li Y, Xu M H, Wang F. et al. Cross-scale assessment of yam waxiness attribute from stress relaxation and fluid mechanics: a distinctive mouthfeel derived from starch matrix. Food Chemistry, 2025, 489: 144896

[7] Deblais A, Hollander E D, Boucon C. et al. Predicting thickness perception of liquid food products from their non-Newtonian rheology. Nature Communications, 2021, 12(1): 6328

[8] Espinosa Y G, Chen J S. Applications of electromyography (EMG) technique for eating studies. In: Chen J S, Engelen L, eds. Food Oral Processing: Fundamentals of Eating and Sensory Perception. Hoboken: John Wiley & Sons, 2012, 289–318

[9] Kohyama K, Hanyu T, Hayakawa F. et al. Electromyographic measurement of eating behaviors for buckwheat noodles. Bioscience, Biotechnology, and Biochemistry, 2010, 74(1): 56–62

[10] Le Révérend B, Saucy F, Moser M. et al. Adaptation of mastication mechanics and eating behaviour to small differences in food texture. Physiology & Behavior, 2016, 165: 136–145

[11] Lee H J, Park Y S, Choi K J. et al. Impact of varying food hardness on mastication/swallowing. Food Science and Biotechnology, 2023, 32(7): 959–967

[12] Oppen D, Weiss J. Oral processing, rheology, and mechanical response: relations in a two-phase food model with anisotropic compounds. Journal of Texture Studies, 2023, 54(6): 808–823

[13] Van Casteren A, Codd J R, Kupczik K. et al. The cost of chewing: the energetics and evolutionary significance of mastication in humans. Science Advances, 2022, 8(33): eabn8351

[14] Boyar M M, Kilcast D. Electromyography as a novel method for examining food texture. Journal of Food Science, 1986, 51(3): 859–860

[15] González R, Montoya I, Cárcel J. Review: the use of electromyography on food texture assessment. Food Science and Technology International, 2001, 7(6): 461–471

[16] Kohyama K, Gao Z H, Watanabe T. et al. Relationships between mechanical properties obtained from compression test and electromyography variables during natural oral processing of gellan gum gels. Journal of Texture Studies, 2017, 48(1): 66–75

[17] Khramova D S, Vityazev F V, Zueva N V. et al. Impact of pectin or xanthan addition to mashed potatoes gelled with κ-carrageenan on texture and rheology, oral processing behavior, bolus properties and in mouth starch digestibility. International Journal of Biological Macromolecules, 2025, 308: 142349

[18] Shiozawa M, Taniguchi H, Hayashi H. et al. Differences in chewing behavior during mastication of foods with different textures. Journal of Texture Studies, 2013, 44(1): 45–55

[19] Zhao Z, Wang Z, Zhu J Y. et al. How differences in rice texture and oral physiology affect sensory preference and eating behavior of consumers: a generalized additive models (GAMs) approach. Food Structure, 2025, 44: 100418

[20] Sato W, Minemoto K, Ikegami A. et al. Facial EMG correlates of subjective hedonic responses during food consumption. Nutrients, 2020, 12(4): 1174

[21] Uehara F, Hori K, Murakami K. et al. Differentiation of feeding behaviors based on masseter and supra-hyoid muscle activity. Frontiers in Physiology, 2020, 11: 618

[22] Wu C C, Jia C Y, Wu M. et al. Mechanisms by which different structures and compositions of exogenous proteins modulate the physicochemical properties of rice starch gels and the textural characteristics of rice noodles. Food Hydrocolloids, 2026, 174: 112390

[23] Huang Y X, Yang X, Cui Y L. et al. Analysis of starch structure, functional properties, and the underlying molecular mechanism in SeaRice86 (SR86). Carbohydrate Polymers, 2025, 366: 123881

[24] Li Y, Zhou Z J, Huang W S. et al. Mechanistic insights into degradation of yam waxiness for freeze-thaw: the interference of ice crystals on dynamic behavior of starch chains. Food Research International, 2026, 227: 118325

[25] Li Y, Li P T, Zhao D Y. et al. Metabolic and starch biosynthetic networks underlying yam waxiness formation revealed by integrated transcriptomics and proteomics. Journal of Agricultural and Food Chemistry, 2025, 73(51): 32846–32861

[26] Wang F, Shen J F, Lu B Y. Intervention mechanism of amphiphilic natural sweeteners on starch chain dynamic behavior: computational and experimental insights. Carbohydrate Polymers, 2025, 349: 122978

[27] Li Y, Ji S Y, Xu M H. et al. Molecular mechanism for the influence of yam starch multiscale structure on the sensory texture of cooked yam. International Journal of Biological Macromolecules, 2024, 271: 132572

[28] Zhu F. Relationships between amylopectin internal molecular structure and physicochemical properties of starch. Trends in Food Science & Technology, 2018, 78: 234–242

[29] Gelders G G, Goesaert H, Delcour J A. Amylose−lipid complexes as controlled lipid release agents during starch gelatinization and pasting. Journal of Agricultural and Food Chemistry, 2006, 54(4): 1493–1499

[30] Ji S Y, Xu T, Liu Y. et al. Investigation of the mechanism of casein protein to enhance 3D printing accuracy of cassava starch gel. Carbohydrate Polymers, 2022, 295: 119827

[31] Zhang Y F, Chen Y, Chen J S. The starch hydrolysis and aroma retention caused by salivary α-amylase during oral processing of food. Current Opinion in Food Science, 2022, 43: 237–245

[32] Wang F, Ma R R, Zhu J L. et al. Design principles and emerging applications of starch-involved superwettable systems. Engineering, 2025, 53: 155–175

[33] Jiang T H, Li X X, Li T L. et al. Ultra-fast wetting of the fresh popcorn. Advanced Functional Materials, 2023, 33(14): 2213036

[34] Yan X D, Xiao H, Chen R Y. et al. Molecular structure of varietal rice starch modulates the supramolecular structure of rice noodle gels: implications for rehydration properties and textural characteristics. Carbohydrate Polymers, 2026, 373: 124606

[35] Farina D, Holobar A, Merletti R. et al. Decoding the neural drive to muscles from the surface electromyogram. Clinical Neurophysiology, 2010, 121(10): 1616–1623

[36] Mao Q, Sun Y H, Hou J M. et al. Relationships of image texture properties with chewing activity and mechanical properties during mastication of bread. International Journal of Food Engineering, 2016, 12(4): 311–321

[37] Jaiswal S, Dhillon B, Sodhi N S. et al. Application of electromyography (EMG) for comprehensive texture evaluation of muffins formulated with incorporation of beetroot and sprouted soy flour. Journal of Food Measurement and Characterization, 2024, 18(6): 5043–5059

[38] Brown W E, Langley K R, Mioche L. et al. Individuality of understanding and assessment of sensory attributes of foods, in particular, tenderness of meat. Food Quality and Preference, 1996, 7(3-4): 205–216

[39] Labouré H, Repoux M, Courcoux P. et al. Inter-individual retronasal aroma release variability during cheese consumption: role of food oral processing. Food Research International, 2014, 64: 692–700

[40] Hennequin M, Allison P J, Veyrune J L. et al. Clinical evaluation of mastication: validation of video versus electromyography. Clinical Nutrition, 2005, 24(2): 314–320

[41] Okubo A, Tsujimura T, Ueha R. et al. The instruction of chewing well affects masticatory activity and bolus flow during rice chewing in healthy humans. Journal of Oral Rehabilitation, 2025, 52(10): 1736–1745

[42] Saito N, Ogawa T, Shiraishi N. et al. Difference in the electromyographic behavior of the masticatory and swallowing muscles during cued versus spontaneous swallowing. Dysphagia, 2024, 39(3): 398–406

[43] Ismael-Mohammed K, Bolívar-Prados M, Laguna L. et al. Exploring texture and biomechanics of food oral processing in fork-mashable dishes for patients with mastication or swallowing impairments. Foods, 2024, 13(12): 1807

[44] Kanda R, Inoue M, Hisajima T. et al. Development and validation of a novel system that combines a new masticatory simulator and analysis method for modeling the human gummy candy masticatory process. Journal of Texture Studies, 2024, 55(1): e12804

[45] Rohatgi B, Ramadoss R, Nitya K. et al. Taste perception and muscular response: EMG based experimental evaluation. Journal of Oral Biology and Craniofacial Research, 2025, 15(3): 472–477

[46] Riente A, Abeltino A, Serantoni C. et al. Evaluation of the chewing pattern through an electromyographic device. Biosensors, 2023, 13(7): 749

[47] Watts C R, Kelly B. The effect of bolus consistency and sex on electrophysiological measures of hyolaryngeal muscle activity during swallowing. Dysphagia, 2015, 30(5): 551–557

[48] Kohyama K, Hayakawa F, Gao Z H. et al. Mouthful size effects on mastication effort of various hydrocolloid gels used as food models. Food Science and Technology Research, 2014, 20(6): 1121–1130

[49] Bourne M C. Food texture and viscosity: concept and measurement. New York: Academic Press, 2002

[50] Peyron M A, Blanc O, Lund J P. et al. Influence of age on adaptability of human mastication. Journal of Neurophysiology, 2004, 92(2): 773–779

[51] Rustagi S, Sodhi N S, Dhillon B. Relationship of electromyography (EMG) masticatory variables with sensory texture and instrumental texture parameters of different textured foods. Journal of Food Measurement and Characterization, 2022, 16(1): 391–399

[52] Sodhi N S, Dhillon B, Pandey V K. et al. Applicability of electromyography (EMG) as a prospective technique for textural evaluation of different types of biscuits. Journal of Agriculture and Food Research, 2024, 16: 101089

[53] Kaur T, Sodhi N S, Dhillon B. et al. Application of electromyography (EMG), an in vivo oral processing analysis technique, for comprehensive texture evaluation of formulated rice products. Journal of Food Measurement and Characterization, 2024, 18(3): 2226–2236

[54] Kohyama K, Sodhi N S, Suzuki K. et al. Texture evaluation of cooked rice prepared from Japanese cultivars using two-bite instrumental test and electromyography. Journal of Texture Studies, 2016, 47(3): 188–198

[55] Chaunier L, Courcoux P, Della Valle G. et al. Physical and sensory evaluation of cornflakes crispness. Journal of Texture Studies, 2005, 36(1): 93–118

[56] da Cunha Rodrigues L, Pereira C D, Furlan G B. et al. Proposing a new crispness index based on mechanical and acoustic properties of food materials. Journal of Food Engineering, 2026, 407: 112845

[57] Taniwaki M, Kohyama K. Mechanical and acoustic evaluation of potato chip crispness using a versatile texture analyzer. Journal of Food Engineering, 2012, 112(4): 268–273

[58] Brown W E, Eves D, Ellison M. et al. Use of combined electromyography and kinesthesiology during mastication to chart the oral breakdown of foodstuffs: relevance to measurement of food texture. Journal of Texture Studies, 1998, 29(2): 145–167

[59] Reyniers S, De Brier N, Ooms N. et al. Amylose molecular fine structure dictates water–oil dynamics during deep-frying and the caloric density of potato crisps. Nature Food, 2020, 1(11): 736–745

[60] Hedjazi L, Guessasma S, Yven C. et al. Preliminary analysis of mastication dynamics and fragmentation during chewing of brittle cereal foods. Food Research International, 2013, 54(2): 1455–1462

[61] Luckett C R, Meullenet J F, Seo H S. Crispness level of potato chips affects temporal dynamics of flavor perception and mastication patterns in adults of different age groups. Food Quality and Preference, 2016, 51: 8–19

[62] Li C F, Ji Y, Li E P. et al. Interactions between leached amylose and protein affect the stickiness of cooked white rice. Food Hydrocolloids, 2023, 135: 108215

[63] Li C. Starch fine molecular structures: the basis for designer rice with slower digestibility and desirable texture properties. Carbohydrate Polymers, 2023, 299: 120217

[64] Kohyama K, Sasaki T, Hayakawa F. Characterization of food physical properties by the mastication parameters measured by electromyography of the jaw-closing muscles and mandibular kinematics in young adults. Bioscience, Biotechnology, and Biochemistry, 2008, 72(7): 1690–1695

[65] Iguchi H, Magara J, Nakamura Y. et al. Changes in jaw muscle activity and the physical properties of foods with different textures during chewing behaviors. Physiology & Behavior, 2015, 152: 217–224

[66] Kohyama K, Sodhi N S, Sasaki T. et al. Effects of milling ratio and water-to-rice ratio on mastication effort for cooked rice measured by electromyography. Journal of Texture Studies, 2014, 45(6): 477–486

[67] Kohyama K, Sawada H, Nonaka M. et al. Textural evaluation of rice cake by chewing and swallowing measurements on human subjects. Bioscience, Biotechnology, and Biochemistry, 2007, 71(2): 358–365

[68] Liu M L, Hu H, Ouyang K F. et al. Effects of Rice, soy, and whey protein additions on quality characteristics and in vitro digestibility of Rice cakes. International Journal of Biological Macromolecules, 2025, 320: 145775

[69] An D, Li H T, Li D S. et al. The relation between wheat starch properties and noodle springiness: from the view of microstructure quantitative analysis of gluten-based network. Food Chemistry, 2022, 393: 133396

[70] Dhillon B, Sodhi N S, Gandotra S. et al. Physico-chemical and textural (sensorial and electromyographic) evaluation of idlis formulated with brown rice and pearl millet flours. Journal of Food Measurement and Characterization, 2020, 14(5): 2898–2906

[71] Zhang Q L, Li X X, Jin Z Y. et al. Enzymatic-chemical modification of cassava starch as a sauce thickener to regulate the balance between processing and edible quality. Carbohydrate Polymers, 2026, 375: 124768

[72] Zhou X L, Chen Y T, Feng P X. et al. Fine structure of starch biomacromolecules and digestibility: the regulative role of amylose and amylopectin in the digestive hydrolysis of starch in rice. Carbohydrate Polymers, 2025, 350: 123040

[73] Diaz-Tay J, Jayasinghe N, Lucas P W. et al. Association between surface electromyography of human jaw-closing muscle and quantified food breakdown. Archives of Oral Biology, 1991, 36(12): 893–898

[74] Matsuyama S, Nakauma M, Funami T. et al. Human physiological responses during swallowing of gel-type foods and its correlation with textural perception. Food Hydrocolloids, 2021, 111: 106353

[75] Ross C F. Sensory science at the human–machine interface. Trends in Food Science & Technology, 2009, 20(2): 63–72

[76] Kaur T, Sodhi N S, Dhillon B. et al. Application of electromyography (EMG) for textural evaluation of raw, steamed and parboiled Basmati rice cultivars (Pusa 1121 and Pusa 1509). Journal of Food Measurement and Characterization, 2023, 17(4): 3228–3239

[77] Casas M J, Kenny D J, Macmillan R E. Buccal and lingual activity during mastication and swallowing in typical adults. Journal of Oral Rehabilitation, 2003, 30(1): 9–16

[78] Park D, Kim Y, Kang H. et al. PECI-Net: bolus segmentation from video fluoroscopic swallowing study images using preprocessing ensemble and cascaded inference. Computers in Biology and Medicine, 2024, 172: 108241

[79] Feng W, Zhang H, Wang R. et al. Modifying the internal structures of steamed rice cakes by emulsifiers for promoted textural and sensory properties. Food Chemistry, 2021, 354: 129469

[80] Saito A, Sato W, Ikegami A. et al. Subjective-physiological coherence during food consumption in older adults. Nutrients, 2022, 14(22): 4736

[81] Grigoriadis A, Johansson R S, Trulsson M. Temporal profile and amplitude of human masseter muscle activity is adapted to food properties during individual chewing cycles. Journal of Oral Rehabilitation, 2014, 41(5): 367–373

[82] Kohyama K, Nakayama Y, Yamaguchi I. et al. Mastication efforts on block and finely cut foods studied by electromyography. Food Quality and Preference, 2007, 18(2): 313–320

[83] Khramova D, Vityazev F V, Zueva N V. et al. Volitional control of chewing frequency affects food oral processing and texture perception. Journal of Oral Rehabilitation, 2025, 52(6): 871–882

[84] Park S, Shin W S. Differences in eating behaviors and masticatory performances by gender and obesity status. Physiology & Behavior, 2015, 138: 69–74

[85] Xu C, Li W, Li S. et al. EA-FUSION: a multimodal emotion recognition model using EEG and facial expression data. Biomedical Signal Processing and Control, 2026, 112: 108629

[86] Sato W, Ikegami A, Ishihara S. et al. Brow and masticatory muscle activity senses subjective hedonic experiences during food consumption. Nutrients, 2021, 13(12): 4216

[87] Sato W, Ishihara S, Ikegami A. et al. Dynamic concordance between subjective and facial EMG hedonic responses during the consumption of gel-type food. Current Research in Food Science, 2025, 10: 101107

[88] Ranawana V, Monro J A, Mishra S. et al. Degree of particle size breakdown during mastication may be a possible cause of interindividual glycemic variability. Nutrition Research, 2010, 30(4): 246–254

[89] Yu H, Tu J X, Hou Y Y. et al. Effect of mastication degree and eating rate on in vitro dynamic gastrointestinal digestion of rice. Food Research International, 2026, 223: 117905

[90] Ranawana V, Henry C J K, Pratt M. Degree of habitual mastication seems to contribute to interindividual variations in the glycemic response to rice but not to spaghetti. Nutrition Research, 2010, 30(6): 382–391

[91] Sun L J, Ranawana D V, Tan W J K. et al. The impact of eating methods on eating rate and glycemic response in healthy adults. Physiology & Behavior, 2015, 139: 505–510

[92] Lv W, Shi S, Chen Z Y. et al. Polysaccharide-protein composite systems in dysphagia-oriented food design: interactions, applications, and additive manufacturing. Food Chemistry, 2026, 522: 149988

[93] Kohyama K, Mioche L, Martin J F. Chewing patterns of various texture foods studied by electromyography in young and elderly populations. Journal of Texture Studies, 2002, 33(4): 269–283

[94] Kohyama K, Mioche L. Chewing behavior observed at different stages of mastication for six foods, studied by electromyography and jaw kinematics in young and elderly subjects. Journal of Texture Studies, 2004, 35(4): 395–414

[95] Kang A J, Kim D K, Kang S H. et al. EMG activity of masseter muscles in the elderly according to rheological properties of solid food. Annals of Rehabilitation Medicine, 2016, 40(3): 447–456

[96] Chen K C, Jeng Y, Wu W T. et al. Sarcopenic dysphagia: a narrative review from diagnosis to intervention. Nutrients, 2021, 13(11): 4043

[97] Rodrigues C A, de Oliveira Melchior M, Magri L V. et al. Is the masticatory function changed in patients with temporomandibular disorder?. Brazilian Dental Journal, 2015, 26(2): 181–185

[98] Cai M D, Cao H W, Long X F. et al. Starch types modulate the texture of dysphagia diets: multiscale characterization and structure-texture relationships of starch-protein-oil composite gels. Food Chemistry, 2026, 499: 147328

[99] Paphangkorakit J, Chaichit R, Khummool K. et al. Effect of biting speed and jaw separation on force used to incise food. Journal of Oral Rehabilitation, 2020, 47(6): 731–735

[100] Tournier C, Grass M, Septier C. et al. The impact of mastication, salivation and food bolus formation on salt release during bread consumption. Food & Function, 2014, 5(11): 2969–2980

[101] Sodhi N S, Dhillon B, Sandhu P S. et al. Formulation of functional liquid diets under IDDSI levels 0 to 2 for dysphagia patients: physico-chemical, sensory, antioxidant and electromyographic oral processing analyses. Dysphagia, 2024, 39(5): 889–904

[102] Alam S A, Pentikäinen S, Närväinen J. et al. Effects of structural and textural properties of brittle cereal foams on mechanisms of oral breakdown and in vitro starch digestibility. Food Research International, 2017, 96: 1–11

[103] Chen Z K, Li X R, Yuan W Q. et al. Nitrogen-regulated amylose structure and complexation during post-ripening: a key to improving rice cooking and eating quality. Carbohydrate Polymers, 2025, 368: 124197

[104] Kohyama K, Ohtsubo K, Toyoshima H. et al. Electromyographic study on cooked rice with different amylose contents. Journal of Texture Studies, 1998, 29(1): 101–113

[105] Miyaoka Y, Ashida I, Tamaki Y. et al. Analysis of masseter activity patterns using T_P values during chewing of foods with different shapes and textural properties. Journal of Texture Studies, 2013, 44(3): 196–204

[106] Pentikäinen S, Sozer N, Närväinen J. et al. Effects of wheat and rye bread structure on mastication process and bolus properties. Food Research International, 2014, 66: 356–364

[107] Kohyama K, Yamaguchi M, Kobori C. et al. Mastication effort estimated by electromyography for cooked rice of differing water content. Bioscience, Biotechnology, and Biochemistry, 2005, 69(9): 1669–1676

[108] Manda Y, Kodama N, Maeda N. et al. Effect of food properties and chewing condition on the electromyographic activity of the posterior tongue. Journal of Oral Rehabilitation, 2019, 46(6): 511–517

[109] Brown W E, Langley K R, Martin A. et al. Characterisation of patterns of chewing behaviour in human subjects and their Influence on texture perception. Journal of Texture Studies, 1994, 25(4): 455–468

[110] Ikenaga N, Yamaguchi K, Daimon S. Effect of mouth breathing on masticatory muscle activity during chewing food. Journal of Oral Rehabilitation, 2013, 40(6): 429–435

[111] Caporizzi R, Derossi A, Devahastin S. et al. Boosting the role of complex food structure on oral breakdown and sweetness perception by digitally designed and 3D printed biscuits. Journal of Food Engineering, 2025, 393: 112499

Rights & permissions

The Author(s) 2026. This article is published by Higher Education Press.