Canonical Analysis Research Articles

Chinese undergraduate students' motivation to learn Korean as a LOTE.

As rates of multilingualism increase, interest in the field of Languages Other Than English (LOTEs) has been growing over the last few years. This study investigated the motivation held by Chinese undergraduate students for learning Korean as a LOTE using Dörnyei's L2 Motivational Self System (L2MSS). In total, 123 subjects responded to the 6-point Likert scale measuring their Korean learning motivation. The collected data were analyzed using SPSS 22. Logistic regression was applied for identifying variables that distinguished the first-year from the second-year learners of Korean, while canonical correlation analysis was used to examine the correlation between two sets of variables, the first set of dependent variables of the ideal L2 self and the ought-to L2 self, and the second set of independent variables of family influence, instrumentality promotion, instrumentality prevention, attitude to learning Korean, cultural interest, attitude toward community and integrativeness. Results showed that variables of family influence, cultural interest, and attitude to learning Korean were statistically significant in distinguishing the first-year from the second-year learners in terms of affective variables. In addition, canonical analysis showed that the dependent variable set of the ideal L2 self and the ought-to L2 self together shared nearly 69% variance with the independent variable set, indicating that the ideal L2 self and the ought-to L2 self together were highly related with these affective variables in the independent variable set. The findings of the current study suggest that more creative Korean language learning activities be adopted to help sustain the high levels of affect among Korean language learners.

Geometric morphometrics and phylogenetic analysis of Delphinium taxa from south of Russia

The leaf blade’s shape of 665 samples of 18 taxa from the genus Delphinium was analyzed using geometric morphometrics method. Within the subgenus Delphinastrum, D. elatum and, to a lesser extent, D. uralense were relatively well separated by canonical analysis, while D. litwinowii, D. pubiflorum, D. duhmbergii, and D. subcuneatumwere not actually separated. Within the subgenus Oligophyllon, D. caucasicum was well separated by this method. The scatter clouds of D. freynii + D. puniceum + D. sergii on the one hand, and D. fedorovii + D. gelmetzicum on the other hand, also separated. D. arcuatum, D. bracteosum, D. crispulum, D. flexuosum, and D. mariae did not separate from each other. The nature of such separation of species groups observed in the space of canonical variables and their relative proximity to each other can be associated with the habitat similarity of their populations driven by ecological or geographical conditions, as well as the influence of hybridization processes in the zone of secondary contact of species. The differences in leaf blade’s shape are mainly related to the width of the segments, the lengths of the central segment of the middle lobe and the undissected part of the leaf blade, the degree of its dissection, the shape of the base and the distance between the lower lobes. Based on the results of DNA sequencing of the intergenic transcribed spacer ITS2, the taxonomic independence of D. puniceum, D. macropogon, D. mariae, D. samurense, and D. pubiflorum is beyond any doubt. The synonymy of D. freynii and D. sergii is supported, with the priority name of D. schmalhausenii. D. cuneatum, D. subcuneatum, D. duhmbergii, and D. litwinowii should be considered as synonyms with the priority name of D. cuneatum. The taxonomic status of D. dictyocarpum, D. elatum, and D. uralense needs further clarification. This is probably related to their easy hybridization with D. cuneatum in sympatric populations within the overlapping areas. Delphinium arcuatum, D. bracteosum, D. caucasicum, D. crispulum, D. elisabethae, D. fedorovii, D. flexuosum, D. gelmetzicum, and D. speciosum are poorly separated, which can also be associated with hybridization processes. They need to be studied in more detail.

Fault Detection for Nonlinear Dynamic Systems With Consideration of Modeling Errors: A Data-Driven Approach.

This article is concerned with data-driven realization of fault detection (FD) for nonlinear dynamic systems. In order to identify and parameterize nonlinear Hammerstein models using dynamic input and output data, a stacked neural network-aided canonical variate analysis (SNNCVA) method is proposed, based on which a data-driven residual generator is formed. Then, the threshold used for FD purposes is obtained via quantiles-based learning, where both estimation errors and approximation errors are considered. Compared with the existing work, the main novelties of this study include: 1) SNNCVA provides a new parameterization strategy for nonlinear Hammerstein systems by utilizing input and output data only; 2) the associated residual generator can ensure FD performance where both the system model and its nonlinearity are unknown; and 3) with consideration of modeling-induced errors, the quantiles are invoked and used to provide a reliable FD threshold in situations where only limited samples are available. Studies on a nonlinear hot rolling mill process demonstrate the effectiveness of the proposed method.

Dynamic State-Space Modeling With Factorial Memories in Temporal Dominance of Sensations, Emotions and Temporal Liking

There are few mathematical models that describe how human perceptual and affective responses evolve with time. We developed a state-space model of the interrelationships based on time-evolving perceptual and affective responses acquired by the temporal dominance (TD) method. We defined and computed the state variables that endow the system with memory using canonical variate analysis. Furthermore, we determined the model parameters on the basis of a cross-validation approach such that the observed and estimated changes in the affective responses are highly correlated. We applied this method to the TD curves of responses to eating strawberries and plum pickles, reported in our previous work. The model describes what happens during food intake in terms of a few state variables that summarize a large number of observable responses reductively, thus helping to make the perceptual and affective dynamics understandable.

Discrimination of milk species based on physico-chemical properties and chemometrics: application of principal component analysis and canonical discriminant analysis

Discrimination of milk species based on physico-chemical properties and chemometrics: application of principal component analysis and canonical discriminant analysis

Comparative Study of Soft Tissue and Hard Tissue Variables Correlation between Two Different Racial groups: A Canonical study

Aims: The purpose of this study was to investigate the correlation between dental, skeletal, and soft tissue variables and compare the results with Saudi study and other previous studies. Material and Method: The sample size consisted of 29 adult Sudanese patients (ages 18 to 25) seeking orthodontic treatment, and 19 variables were examined. The error of the method was assessed using Student's t-test, and canonical correlation analysis was used to analyze the correlation between soft and hard tissues. Results: The study found that the Sudanese population had higher mean values in dental and soft tissue variables compared to the Saudi population, while similar mean values were found in the skeletal sagittal relationship, except for the vertical relationship, where the Saudi population had a higher tendency for an open basal configuration. The canonical analysis revealed that the upper and lower incisors' inclinations had a strong negative loading, while the nasolabial angle and lower lip thickness and position had moderate loading. Conclusion: These findings show the impact of these variables which can help the clinician in diagnosis, prediction and assessment of post treatment changes. Additionally, the multivariate statistical analysis extracted a clinically significant association between soft tissue and hard tissue.

Assessment of wing geometric morphometrics of urban Culex quinquefasciatus (Diptera: Culicidae) populations

Culex quinquefasciatus is a cosmopolitan species distributed throughout tropical and subtropical areas of the world. The species is of great epidemiological importance as it is responsible for vectoring the causative agent of lymphatic filariasis and several arboviruses, including West Nile virus. Wing geometric morphometrics has been widely used to assess phenotypic variations in mosquito species. Here, we hypothesize that Cx. quinquefasciatus populations in urban parks in the city of São Paulo, Brazil, have been subjected to anthropogenic selective pressures that are responsible for driving their ecology and behavior. Mosquitoes were collected by CDC traps in five municipal parks in the city of São Paulo. Eighteen anatomical landmark coordinates on each female right wing were digitized. Canonical variate analysis, wireframe graphs, cross-validated reclassification tests and the neighbor-joining method were used to assess phenotypical dissimilarity in wing shape between populations. Centroid size was calculated to assess differences in wing size between populations, which can result from different environmental conditions during immature mosquito development. Moderately heterogeneous wing shape and wing size patterns were found in the populations analyzed, indicating that selective pressures in the urban environment are affecting the wing patterns of Cx. quinquefasciatus populations in the city of São Paulo, Brazil.

Analisis intertekstual Roma 9:33 Suatu Pendekatan Memori Semantik Pada Teologi Paulus

Romans 9:33 is known as a very complicated verse with Old Testament intertextual quotations from Isaiah 8:14 and 28:16 in splits and crosses. The use of allusions as key words such as "Zion" and "stone" requires analytical methods and specific rules so as not to be biased in semantic interpretation. Even the change of “disturbed” (Isa 28:16 KJV) to “be ashamed” (Rom 9:33 KJV) has meaning. The inerrancy of the Scriptures can be questioned if it is called just a variation or a scribal error, because there is no scribal doubtness in the textual critics. The research aims to prove Paul's historical canonical semantic intertextual writing technique in Romans 9:33 as a key verse explaining theology justified by faith to the Roman plural congregation. Qualitative research through canonical intertextual analysis is carried out using a semantic historical memory approach through multidisciplinary discourses such as cognitive psychology and architectural archeology. The results of the study show that Romans 9:33 was written using the synchronic multiple contexts semantic memory method which is capable of explaining with precision the theology justified by faith, both to the Jewish and non-Jewish congregations in Rome. The result in this study also provides new afirmation of the dynamics of intertextuality poststructuralism in a canonical context for modern hermeneutics.

Photosynthetic Assimilation of the Guava (Psidium guajava) cv. Paluma under Different Pruning and Fruit Thinning Intensities

In guava plants, production pruning can be performed twice a year, and the return of growth is dependent on the physiological responses that are altered by the different cultivation environments and adopted management. From this perspective, this study aimed to characterize the photosynthetic dynamics of guava plants influenced by different pruning and fruit thinning intensities during two growing seasons in the region of Currais, Piauí, Brazil. The plants were distributed in a randomized block design with a factorial arrangement (3 × 3 × 2) consisting of three pruning intensities (short, medium, and long) and three fruit thinning intensities (0, 10, and 20%) during two growing seasons. The data were subjected to a cluster analysis and canonical discriminant analysis to discriminate treatment groups based on the variables. Through a cluster analysis for the evaluated treatments, it was possible to split the two pruning seasons into five different groups clustered for the first pruning season and the second pruning season. The highest assimilation values were observed in the first pruning season and especially in plants that received short pruning with 0% fruit thinning, medium pruning with 10% and 20% fruit thinning, and long pruning with 10% fruit thinning. Through the graphic representation of the canonical discriminant analysis, the first two variables explained 93.40% of the total variance contained in the nine original variables. The highest means of ambient PAR, transpiration, leaf temperature, internal carbon, and ambient temperature were observed in the second pruning season and in plants that received short pruning with 10% and 20% fruit thinning, medium pruning with 0% and 20% fruit thinning, and long pruning with 0%, 10%, and 20% fruit thinning favors a higher photosynthetic accumulation in guava plants. We observed a multiplicity of responses; however, short pruning with 10% thinning should be considered for both seasons.

Elements in Soft Tissues of the Young Mediterranean Mussel Mytilus galloprovincialis Lam. 1819 Collected in Sevastopol Bay (Crimea, Black Sea): Effects of Age, Sex, Location, and Principal Morphometric Parameters.

Although the mussel Mytilus galloprovincialis has been known for decades as an excellent bioindicator of trace element pollution in the marine environment, there is still no information on the effects of a suite of its principal morphometric parameters and age on trace element levels in soft tissues. In this work, using inductively coupled plasma mass spectrometry, we studied the contents of 72 elements in soft tissues of M. galloprovincialis aged 0.5-4, which were sampled at three stations within a relatively small water body, Sevastopol Bay. Significant effects of age and sampling location on the element contents and soft tissue dry-to-wet weight ratio were discovered. The effects of sex were not significant. It was presumed for the first time that the decrease in element content in the soft tissues of young mussels can be associated with the decrease in physiological needs for elevated contents of essential elements and intracellular water with age. Combinations of six principal morphometric parameters showed that a function of as few as three parameters (soft tissue dry weight, whole mollusk weight, and shell height, with by far the greatest contribution of the dry-to-total weight ratio) formed significant correlations with the contents of the largest possible number of elements (69-88% of the total number). For the first time, it was shown that linear discriminant analysis and canonical analysis of principal coordinates can be successfully used for tracing the exact origin of mussel samples within such a small water area. Canonical analysis of principal coordinates proved superior in the correct classification of the samples.

Differential impacts of ionizing radiation on a sexually dimorphic trait in male and female Acheta domesticus

Purpose In many Orthopteran species, including crickets, forewings exhibit substantial sexual dimorphism driven by sexual selection. In the cricket, Acheta domesticus, females are the ‘choosy’ sex and males exhibit multiple sexual signals to attract and successfully mate. Male forewings have highly specialized structures critical for acoustic signaling and mating. In contrast, female forewings currently serve no known purpose in this flightless species. Forewings also differ morphologically with male forewings containing complex acoustic producing and resonating regions and females lacking any defined structures. Due to their importance to mating as well as their structural complexity, impacts of environmental stress that target cricket forewing development may therefore have more severe consequences in males than females. Here, we investigate the sensitivity of a sexually dimorphic trait, forewing morphology, to an early life environmental stressor. Materials and Methods We applied ionizing radiation (0-–27.8 Gy) as a stressor as dose can be precisely applied as well as its relevance in both environmental contamination and use in the Sterile Insect Technique. Results A canonical variate analysis indicated that wing shape was significantly altered in males at all doses; .58 Gy, 2.3 Gy, 4.6 Gy, 16.2 Gy, and 23.2 Gy. In females, shape was significantly altered at 27.8 Gy and 23.2 Gy groups and to a lesser extent at .58 Gy and 16.2 Gy. Linear regression analysis of centroid size indicated a dose dependent decline in wing size in both sexes, with males exhibiting more decline. Fluctuating asymmetry, a measure of environmental sensitivity, revealed that males were more sensitive to shape changes due to stress than females. This difference in sensitivity is likely due to the complexity of male forewings. Conclusion These results expand understanding of sex dimorphism in stress responses and sensitivity to ionizing radiation.

Age estimation from mandibles in Malay: A 2D geometric morphometric analysis

ObjectivesIn this study, the sizes and forms of mandibles in various age groups of the Malay population were measured and compared. MethodsGeometric morphometric (GM) analysis of mandibles from 400 dental panoramic tomography (DPT) specimens was conducted. The MorphoJ program was used to perform generalized Procrustes analysis (GPA), Procrustes ANOVA, principal component analysis (PCA), discriminant function analysis (DFA), and canonical variate analysis (CVA). In the tpsDig2 program, the 27 landmarks were applied to the DPT radiographs. Variations in mandibular size and form were categorized into four age groups: group 1 (15–24 years), group 2 (25–34 years), group 3 (35–44 years), and group 4 (45–54 years). ResultsThe diversity in mandibular shape among the first eight principal components was 81%. Procrustes ANOVA revealed significant shape differences (P < 0.001) among age groups. Mahalanobis distances indicated substantial differences among all age groups; group 1 and group 4 scored highest, at 2.114. The ranges for the cross-validation and discriminant function tests were 90–72% and 81–49%, respectively. ConclusionGM analysis through radiography is a simple, non-invasive, and non-destructive method of estimating age by using the mandible. GM analysis is unique because it can visualize the changes in mandible shape among age groups. This method should aid in age identification in forensic odontology investigations.

Colour descriptors for plant-based milk alternatives discrimination.

Colour characteristics of plant-based milk alternatives (PBMAs: almond, coconut, cashew, oat, soy) were monitored during long-term storage to select suitable descriptors for PBMAs differentiation. All colour descriptors evaluated varied depending on the raw material used in the plant-based milk alternative production. Long-term storage of plant-based beverages resulted in slightly noticeable (0.5-1.5) and noticeable (1.5-3.0) colour changes. Based on all colour descriptors, an absolute differentiation of PBMAs according to the type of raw material and storage time was achieved using canonical discriminant analysis. The results also indicate the possibility of using colour descriptors to detect the addition of honey to these products. Statistical analysis identified yellowness, browning index and lightness as the most discriminating parameters.

Identifying Neural Signatures of Dopamine Signaling with Machine Learning.

The emergence of new tools to image neurotransmitters, neuromodulators, and neuropeptides has transformed our understanding of the role of neurochemistry in brain development and cognition, yet analysis of this new dimension of neurobiological information remains challenging. Here, we image dopamine modulation in striatal brain tissue slices with near-infrared catecholamine nanosensors (nIRCat) and implement machine learning to determine which features of dopamine modulation are unique to changes in stimulation strength, and to different neuroanatomical regions. We trained a support vector machine and a random forest classifier to decide whether the recordings were made from the dorsolateral striatum (DLS) versus the dorsomedial striatum (DMS) and find that machine learning is able to accurately distinguish dopamine release that occurs in DLS from that occurring in DMS in a manner unachievable with canonical statistical analysis. Furthermore, our analysis determines that dopamine modulatory signals including the number of unique dopamine release sites and peak dopamine released per stimulation event are most predictive of neuroanatomy. This is in light of integrated neuromodulator amount being the conventional metric used to monitor neuromodulation in animal studies. Lastly, our study finds that machine learning discrimination of different stimulation strengths or neuroanatomical regions is only possible in adult animals, suggesting a high degree of variability in dopamine modulatory kinetics during animal development. Our study highlights that machine learning could become a broadly utilized tool to differentiate between neuroanatomical regions or between neurotypical and disease states, with features not detectable by conventional statistical analysis.

The structure of acoustic voice variation in bilingual speech.

When a bilingual switches languages, do they switch their voice? Using a conversational corpus of speech from early Cantonese-English bilinguals (n = 34), this paper examines the talker-specific acoustic signatures of bilingual voices. Following the psychoacoustic model of voice, 24 filter and source-based acoustic measurements are estimated. The analysis summarizes mean differences for these dimensions and identifies the underlying structure of each talker's voice across languages with principal component analyses. Canonical redundancy analyses demonstrate that while talkers vary in the degree to which they have the same voice across languages, all talkers show strong similarity with themselves, suggesting an individual's voice remains relatively constant across languages. Voice variability is sensitive to sample size, and we establish the required sample to settle on a consistent impression of one's voice. These results have implications for human and machine voice recognition for bilinguals and monolinguals and speak to the substance of voice prototypes.

Thermoregulatory capacity of goat kids from birth to weaning in a dry hot climate.

This study aimed to evaluate the thermoregulatory capacity and performance of Saanen goat kids from birth to weaning in a hot climate. Twelve newborn males and female goat kids with an initial body weight of 4.17 ± 0.81 kg were used. Physiological responses, climatic variables, and biometric traits data were collected. Univariate and multivariate analysis techniques were used. Heart rate (HR) was high up to the 6th week of life, with a reduction from the 7th week on (P < 0.001). Rectal temperature (RT) was lower in the first 2 weeks (P < 0.001), with an increase and stabilization occurring in the 7th and 8th weeks. Coat surface temperature (ST) was more activated from the 5th week onwards (P < 0.001). Body weight (BW) and withers height (WH) were higher in later weeks of the calving phase with a linear effect (P < 0.001). The first principal component demonstrated the relationship of sensible heat dissipation × body area of the goat kids; the second component shows the relationship of meteorological data with RT, having a positive relationship between RT with RH and negative with AT, and the third component points to the association of RR and HR. Of the animals, 81.3% were correctly classified in their group of origin in discriminant canonical analysis, with emphasis on the classification of the kids in the 1st-2nd and 3rd-4th weeks (classification percentage Ʃ = 95.8%). It is concluded that (i) newborn kids activate latent mechanisms to maintain their homoeothermic during the first 2 weeks of life, and as they grow, they use sensitive heat loss processes, especially from the 5th week of life onwards and (ii) male and female goats do not show sexual dimorphism effect on body performance and body morphometric measurements up to 60 days of life.

The use of morphometric characters to discriminate populations of Heterorhabditis bacteriophora (Nematoda: Heterorhabditidae).

In a survey of entomopathogenic nematodes (EPNs) in the northwest of Iran, eleven isolates of Heterorhabditis bacteriophora were recovered mainly from the soil samples of apple orchards. The isolates were identified morphologically and molecularly. To unravel intraspecific variation, morphometric characters of different life stages, including hermaphrodite, female, male and infective juvenile (IJ), were evaluated to see efficiencies in discrimination of H. bacteriophora populations isolated from relatively limited geographical regions. Morphometric data were subjected to principal component analysis (PCA), canonical analysis (CAN), and also cluster analysis. Significant differences in morphometrics were recorded between the studied populations. The results of the PCA on IJ variables showed that the first two PC account for more than 99% of the total variance. Variables including distance from anterior end to the base of the pharynx (Es) and distance from anterior end to nerve ring (Nr) showed the highest factor loading. Cluster analysis put the studied populations in five well-defined groups. Resulted PC1 and PC2 for females account 63% of the total variance. Females variables include body length (L), distance from anterior end to excretory pore (EP), length divided by width (A), length divided by Es (B), length divided by tail length (C), and anal body width (Abw), tail length (TL), EP divided by TL ×100 (E) were effective in discrimination of the populations into three distinct groups. The first two PC from hermaphrodite variables accounted for 66% of the total variance. Variables include L, A, B, and C had efficient factor loading on PC1, and PC2. Cluster analysis of hermaphrodite morphometrics yielded five distinct groups among populations. For male variables PC1, and PC2 accounted 60.04% of the total variance, and body width, Abw, TL and, E were the most efficient in PC1 and PC2 was affected efficiently by Es and EP. The results of PCA showed that the highest discrimination among the studied H. bacteriophora population occurred with female variables.

Portable Instruments Based on NIR Sensors and Multivariate Statistical Methods for a Semiautomatic Quality Control of Textiles

Near-infrared (NIR) spectroscopy is a widely used technique for determining the composition of textile fibers. This paper analyzes the possibility of using low-cost portable NIR sensors based on InGaAs PIN photodiode array detectors to acquire the NIR spectra of textile samples. The NIR spectra are then processed by applying a sequential application of multivariate statistical methods (principal component analysis, canonical variate analysis, and the k-nearest neighbor classifier) to classify the textile samples based on their composition. This paper tries to solve a real problem faced by a knitwear manufacturer, which arose because different pieces of the same garment were made with “identical” acrylic yarns from two suppliers. The sweaters had a composition of 50% acrylic, 45% wool, and 5% viscose. The problem occurred after the garments were dyed, where different shades were observed due to the different origins of the acrylic yarns. This is a challenging real-world problem for two reasons. First, there is the need to differentiate between acrylic yarns of different origins, which experts say cannot be visually distinguished before garments are dyed. Second, measurements are made in the field using portable NIR sensors rather than in a controlled laboratory using sophisticated and expensive benchtop NIR spectrometers. The experimental results obtained with the portable sensors achieved a classification accuracy of 95%, slightly lower than the 100% obtained with the high-performance laboratory benchtop NIR spectrometer. The results presented in this paper show that portable NIR sensors combined with appropriate multivariate statistical classification methods can be effectively used for on-site textile quality control.

Effect of sainfoin proanthocyanidins on milk fatty acids from ewes rearing suckling lambs

Proanthocyanidins (PAC) can modulate the fatty acid (FA) profile of animal products and make them healthier for human consumption, but their effects are highly variable depending on several factors such as PAC chemical structure or dose. The present experiment aimed to evaluate the effect of PAC on the milk FA profile of Rasa Aragonesa ewes fed fresh sainfoin (PAC-containing forage legume) during the rearing period of suckling lambs (4 weeks postlambing). Twenty lactating ewes rearing a single male lamb were fed fresh sainfoin ad libitum plus 200 g/d of barley. Half the ewes were orally dosed with 100 mL of water (Sainfoin Group; n = 10) and the other half with polyethylene glycol (50 g PEG4000/100 mL water, Sainfoin + PEG Group; n = 10) to block PAC effects. Sainfoin and milk samples were collected weekly to determine their FA profile by gas chromatography. Fresh sainfoin contents of C18:3n-3 decreased from week 1 to week 2, C16:0 and C18:0 increased from week 1 to week 3, and C18:2n-6 did not change. Regarding milk FA, there were minor effects of PAC on milk-saturated FA. During the whole study, the presence of PAC increased C18:0, C18:2n-6, C18:3n-3 and total polyunsaturated FA (PUFA) n-6 and n-3, and decreased C18:1 t11, branched- and odd-chain FA. However, the decrease of milk concentrations of trans-monounsaturated FA, C18:1 t10, and total conjugated linoleic acid (CLA) and the increase of total PUFA due to the presence of PAC occurred only in week 1, while CLA c9,t11 was lower during weeks 1 and 2. The canonical analyses confirmed the differences between treatments in the FA profile of milk. Overall, the use of fresh sainfoin in the diet of lactating ewes resulted in a beneficial modification of the concentration of several milk FAs, suggesting some changes in ruminal biohydrogenation.

60 Bite and mouth shape variations assessment during different slow-feeding management in ponies: A geometric morphometric evaluation

Previous studies have highlighted the influence of different slow-feeding devices on behavior, intake rate and body postures of stabled equines. There is still, however, a lack of knowledge on how feeding devices change the biting mechanism and ultimately chewing behavior. The aim of our study was to evaluate the variation of bite and mouth shape in ponies fed the same forage by different feeding methods. Nine ponies belonging to 2 breed types - Shetland (n = 5, SH) and Welsh Cob (n = 4, WC) types - were used (mean ± SD: 12 ± 4 years and a BCS:7.6 ± 0.7/9). Breed type designation was assigned according to owner information and body morphometric measurements. A first cut meadow hay was offered to SH and WC ponies in 4 ways following a Latin square design: 3 kg of hay fed on the ground (G); or a small-holed hay net (3.5 × 3.5 cm holes) filled with 3 kg (Haynet full fill, HNF) or 1 kg (Haynet light fill, HNL) of hay; or a newly developed slow feeder (Haybox, HB) containing 3 kg of hay. The 2 hay nets were optimally placed, according to previous work, with the hay net base at the same height as the pony's elbow. Each pony was video recorded 3 times per feeding method and from the video at least 10 still frames/pony/feeding method were selected. A specific point configuration comprising 9 landmarks was developed and geometric morphometrics was applied to study variation in bite and mouthshape. Data were statistically analyzed using principal component analysis (PCA) and canonical variate analysis (CVA). When ponies were fed on the ground (G), the CVA showed the lowest variation (35.7%) in the bite and mouth shape, suggesting a similar biting mechanism despite the morphometric differences between the 2 breed types. In contrast, the use of the slow feeding devices – HB, HNF, HNL – showed higher values of shape variation between the breeds. Interestingly, the highest variation was reported when using the HNL (83.2%), followed by the HB (78.3%) and by the HNF (76.7%), highlighting that each feeding device made the animal alter their biting mechanism. These results were confirmed visually through PCA analysis, demonstrating that the features of the feeding device influenced the mouth shape. Our study suggested that the use of slow-feeding devices, of different size and capacity, not only can alter the intake rate of ponies but can also affect their way of feeding in terms of bite and mouth shape regardless of their morphology. This should be taken into account in further studies on slow feeding management, with the aim to fully understand the effects of different devices on feed intake mechanics.