Discrimination Experiments Research Articles

Wearable High-Resolution Haptic Display Using Suction Stimuli to Represent Cutaneous Contact Information on Finger Pad.

A high-resolution haptic display that reproduces tactile distribution information on the contact surface between a finger and an object realizes the presentation of the softness of the object and the magnitude and direction of the applied force. In this article, we developed a 32-channel suction haptic display that can reproduce tactile distribution on fingertips with high resolution. The device is wearable, compact, and lightweight, thanks to the absence of actuators on the finger. A FE analysis of the skin deformation confirmed that the suction stimulus interfered less with adjacent stimuli in the skin than when pressing with positive pressure, thus allowing more precise control of local tactile stimuli. The optimal layout with the least error was selected from three configurations dividing 62 suction holes into 32 ports.The suction pressures were determined by calculating the pressure distribution by a real-time finite element simulation of the contact between the elastic object and the rigid finger. A discrimination experiment of softness with different Young's modulus and its JND investigation suggested that the higher resolution of the suction display improved the performance of the softness presentation compared to a 16-channel suction display previously developed by the authors.

Research on Apple Origins Classification Optimization Based on Least-Angle Regression in Instance Selection

Machine learning is used widely in near-infrared spectroscopy (NIRS) for fruit qualification. However, the directly split training set used contains redundant samples, and errors may be introduced into the model. Euclidean distance-based and K-nearest neighbor-based instance selection (IS) methods are widely used to remove useless samples because of their accessibility. However, they either have high accuracy and low compression or vice versa. To compress the sample size while improving the accuracy, the least-angle regression (LAR) method was proposed for classification instance selection, and a discrimination experiment was conducted on a total of four origins of 952 apples. The sample sets were split into the raw training set and testing set; the optimal training samples were selected using the LAR-based instance selection (LARIS) method, and the four other selection methods were compared. The results showed that 26.9% of the raw training samples were selected using LARIS, and the model based on these training samples had the highest accuracy. Thus, the apple origin classification model based on LARIS can achieve the goal of high accuracy and compression and provide experimental support for the least-angle regression algorithm in classification instance selection.

Perceptibility of programmable softness displays using magnetorheological elastomers

While often focused on our visual system, adding touch to VR/AR environments can help render more immersive, richer user experiences. One important touch percept to render is compliance, or ‘softness.’ Herein, we evaluate the perceptibility of soft, magnetorheological elastomers (MRE) in bare-finger interactions. Such materials can be reprogrammed to distinct states of compliance. We fabricated MRE samples over elastic moduli from 23–173 kPa and measured that small 0.25 T magnetic fields increased modulus by 10–60 kPa. MRE interfaces less and more compliant than finger skin were evaluated in discrimination experiments with and without a magnetic field. The results indicate changes in modulus of 11 kPa are required to reach a 75% threshold of discrimination, although greater differences are required when an MRE’s elasticity is about the same as skin. The perceptual results with these magnetically-induced materials are similar to those with non-actuated, solid silicone-elastomers that mimic naturalistic interactions.

Do choice tasks and rating scales elicit the same judgments?

Discrete choice (DCE) and rating scale experiments (RSE) are commonly applied procedures for eliciting preference judgments in a plethora of applied settings such as consumer choices, health care, and transport economics. An almost universal assumption is that actual “ground truth” preferences do not depend on which elicitation procedure is used. It is usually not possible to test this assumption, because typical studies feature response options for which there is no objectively correct response. To make progress on testing this assumption, we conducted a perceptual discrimination experiment where response options varied on a single attribute – stimulus saturation level – with a known objectively correct response. We had the same participants complete both a choice task (CT) and rating scale (RS) version of the experiment, allowing a direct examination of the assumption of a common representation. Our CT featured many characteristics that define a DCE, however, in order to have a known objectively correct response, it also differed in a few important ways. To test the assumption of a common representation, we developed a cognitive model with a response mechanism for both CT and RS. This enabled us to compare a model version that featured one shared latent stimulus representation across CT and RS versus a version which featured separate representations. Our results support the assumption that a single internal state supports both CT and RS responses, and also suggest that the CT method might provide more sensitive measurement of internal states than the RS method.

DLPFC-PPC-cTBS effects on metacognitive awareness

BackgroundNeuroimaging and lesion studies suggested that the dorsolateral prefrontal and posterior parietal cortices mediate visual metacognitive awareness. The causal evidence provided by non-invasive brain stimulation, however, is inconsistent. Objective/hypothesisHere we revisit a major figure discrimination experiment adding a new Kanizsa figure task trying to resolve whether bilateral continuous theta-burst transcranial magnetic stimulation (cTBS) over these regions affects perceptual metacognition. Specifically, we tested whether subjective visibility ratings and/or metacognitive efficiency are lower when cTBS is applied to these two regions in comparison to an active control region. MethodsA within-subjects design including three sessions spaced by one-week intervals was implemented. In each session, every participant was administered bilateral cTBS to either prefrontal, control or parietal cortices. Two concurrent tasks were performed, a real and an illusory figure task, stabilising objective performance with use of an adaptive staircase procedure. ResultsWhen performing the replicated task, cTBS was found insufficient to disrupt neither visibility ratings nor metacognitive efficiency. However, with use of Kanizsa style illusory figures, cTBS over the dorsolateral prefrontal, but not over the posterior parietal cortex, was observed to significantly diminish metacognitive efficiency. Conclusion(s)Real and illusory figure tasks demonstrated different cTBS effects. A possible explanation is the involvement of the prefrontal cortex in the creation of expectations, which is necessary for efficient metacognition. Failure to replicate previous findings for the real figure task, however, cannot be said to support, conclusively, the notion that these brain regions have a causal role in metacognitive awareness. This inconsistent finding may result from certain limitations of our study, thereby suggesting the need for yet further investigation.

Sufficient Time-Frequency Resolution for Reproducing Vibrotactile Sensation.

In vibrotactile stimuli, it is essential to reproduce realistic tactile sensations to enhance the immersiveness of applications. To reproduce more realistic tactile experiences, various tools have been proposed to fine-tune and design vibrotactile sensations. Considering the situation where users adjust parameters manually, providing tactile sensations with fewer parameters is desirable. This study examines the coarsest resolution in the time and frequency dimensions necessary to present tactile sensations as realistic as vibrations recorded by the sensor. Time and frequency are fundamental parameters to express vibrations as a spectrogram, and we considered it important to investigate how much coarser the resolution could be without changing perception. We focus on the textural vibrations and the preliminary experiment compared actual texture vibrations with the reconstructed vibration as coarse as possible in the frequency dimension. The result showed that the frequency resolution above 172 Hz makes it difficult to distinguish between the vibrations. The main experiment, a similar discrimination experiment, verified the time resolution using the averaging filter of vibration intensity over time. The results indicate that with the update interval set to 30 ms, the discrimination rate compared to the original vibration is approximately 60%. This percentage is below the chance level of 75%, indicating that distinguishing between the two is difficult. Based on our experiments, it is necessary to have a frequency resolution of at least 172 Hz and a time resolution that updates intensity at a rate of 30 fps or higher to recreate tactile sensations comparable to actual vibrations.

L1 variation and L2 acquisition: L1 German /eː/-/ɛː/ overlap and its effect on the acquisition of L2 English /ɛ/-/æ/.

A person's first language (L1) affects the way they acquire speech in a second language (L2). However, we know relatively little about the role different varieties of the L1 play in the acquisition of L2 speech. This study focuses on German (L1) learners of English (L2) and asks whether the degree to which German speakers distinguish between the two vowels /eː/ and /ɛː/ in their L1 has an impact on how well these individuals identify /æ/ and discriminate between the two English vowels /ɛ/ and /æ/. These two English vowels differ in both vowel quality and duration (/æ/ is longer than /ɛ/). We report on an identification and a discrimination experiment. In the first study, participants heard a sound file and were asked to indicate whether they heard "pen" or "pan" (or "pedal" or "paddle"). The stimuli differed from each other in terms of both vowel quality (11 steps on a spectral continuum from an extreme /æ/ to an extreme /ɛ/) and duration (short, middle, long). In the second study, participants had to signal whether two sound files they were exposed to differed from each other. We modeled the percentage of /æ/ ("pan," "paddle") selection (identification task only, binomial logistic regression), accuracy (discrimination task only, binomial logistic regression), and reaction time (identification and discrimination tasks, linear mixed effects models) by implementing the German Pillai score as a measure of vowel overlap in our analysis. Each participant has an individual Pillai score, which ranges from 0 (= merger of L1 German /eː/ and /ɛː/) to 1 (=maintenance of the contrast between L1 German /eː/ and /ɛː/) and had been established, prior to the perception experiments reported here, in a production study. Although the findings from the discrimination study remain inconclusive, the results from the identification test support the hypothesis that maintaining the vowel contrast in the L1 German leads to a more native-like identification of /æ/ in L2 English. We conclude that sociolinguistic variation in someone's L1 can affect L2 acquisition.

ISEE: A Wearable Image-Sound Translation System for Partially Sighted People

As a growing number of blinds, many blind aid systems which use the way of converting image to sound or vibration have been designed. However, phonetic prompt and vibration are very limited in information expression. Therefore, a system called Image-Sound Eye-substitution Equipment (ISEE) is designed in this paper. This system consists of binocular camera, Raspberry Pi, and stereo headsets. It can convert image to sound by mapping each pixel into an independent sound source based on sensory substitution technology and enable blinds to browse pictures, appreciate scenery, and avoid obstacles for its high-density information expression. Its core of data processing design is the Sensory Substitution Technology-based Image-to-Sound Transform (SST-IST) algorithm. The SST-IST mainly consists of 3 steps, i.e., the image reshape, image-to-sound conversion matrix generation, and image-to-sound transmission. First, a single-frequency pitch of pixel is generated, followed by wave interception and localization to obtain independent sound clips to express pixel information. Second, the sound chips based on Head Related Transfer Functions (HRTF) are stereo processed. Third, a series of single-frequency sound chips are accumulated with weights to generate a complete sound. Extensive experiments show that our SST-IST can restore image information well and the sound restore rate can achieve 73.0%. Also, it is universally adaptive and robust to the changes in ambient light and noises within a certain range. The auditory discrimination experiments show that, with 4-5 hours training, the users can recognize basic graphics by using ISEE system.

Assessing the time course and magnitude of different forms of attentional priming.

Priming of attentional selection involves speeded selection of task-relevant visual search items when search stimuli remain constant between trials. Various paradigms involving different features have been used to study the nature of this priming. The tasks differ greatly in difficulty and the neural mechanisms involved, raising the question of how easily priming on one feature dimension can be used to draw conclusions about priming on another. Here, this was addressed by contrasting time courses and relative sizes of priming effects for the repetition of a lower-level and higher-level feature (color vs. facial expression). Priming was tested in two odd-one-out search tasks, one involving discrimination (experiments 1A and 1B), the other a present/absent judgment (experiments 2A and 2B). The main question was how similar the size and temporal profiles of priming are for the two features. The sizes of the priming effects were very different for color and expression and color priming effects lasted for much longer than expression priming (measured with memory kernel analyses), suggesting that the mechanisms behind the effects differ in their operational principles. Different forms of priming should only be compared with great caution and priming seems to occur at many levels of processing. Priming should be thought of as a general principle of perceptual processing.

On the auditory identifiability of Asian American identity in speech: The role of listener background, sociolinguistic awareness, and language ideologies

The current study examined the auditory identifiability of Asian American ethnoracial identity, including the role of listener characteristics and ideologies. Results of an identification experiment showed that the overall accuracy of ethnoracial identification on (East and Southeast) Asian talkers was low, but not the lowest among talker groups and not significantly different from accuracy on Black talkers. There were also significant effects of listeners' ethnoracial identity, gender, and linguistic chauvinism (i.e., disfavoring linguistic diversity in the US). In particular, being Asian or a woman was associated with a higher likelihood of accuracy, whereas greater linguistic chauvinism was, to an extent, associated with a lower likelihood of accuracy. Results of a discrimination experiment additionally showed an effect of listeners' awareness of ethnoracially-based language variation: having this awareness was associated with a higher likelihood of accuracy on discrimination trials with one or more Asian talkers. Taken together, these findings converge with previous results showing an effect of the listener's background on ethnoracial perception and further implicate the listener's sociolinguistic awareness and ideologies.

The sound of timbre reigns: Auditory sequence discrimination in 3–5-year-old children

ABSTRACT How do young children apprehend the sounds in their environments? The current study elucidates a recent finding that 3–5-year-olds have greater sensitivity to timbre – non-pitch-related spectral and dynamic attributes that differentiate sounds – than to pitch contour. This finding is somewhat surprising given that pitch contour underlies much of linguistic prosody and music recognition. Further, it represents a gap in scientific knowledge about nonspeech auditory perception, which is commonly linked to, and used to remediate, speech perception. Two hypotheses are explored: the temporal order hypothesis (that pitch contour discrimination is more difficult because it requires encoding the order of pitches, while timbre discrimination does not); the perceptual distance hypothesis (that previously tested timbres were more auditorily distant to children than pitches were, and thus that greater perceptual distance even within a dimension should lead to better discrimination). Three same-different discrimination experiments with 3–5-year-old children suggested that temporal order variation makes timbre change detection harder, but does not affect pitch change detection, and that more-distant pitch changes are better detected (CG vs. GC is easier than CD vs. DC), with hints at such an effect for timbre. Findings raise the possibility of a trade-off such that auditory perceptual distance aids detection only up to a point, past which sounds cohere less well, impeding apprehension of temporal order. Implications for auditory and spoken language development are discussed.

Discrimination tasks in simulated low-dose CT noise.

This study reports the results of a set of discrimination experiments using simulated images that represent the appearance of subtle lesions in low-dose computed tomography (CT) of the lungs. Noise in these images has a characteristic ramp-spectrum before apodization by noise control filters. We consider three specific diagnostic features that determine whether a lesion is considered malignant or benign, two system-resolution levels, and four apodization levels for a total of 24 experimental conditions. The goal of the investigation is to better understand how well human observers perform subtle discrimination tasks like these, and the mechanisms of that performance. We use a forced-choice psychophysical paradigm to estimate observer efficiency and classification images. These measures quantify how effectively subjects can read the images, and how they use images to perform discrimination tasks across the different imaging conditions. The simulated CT images used as stimuli in the psychophysical experiments are generated from high-resolution objects passed through a modulation transfer function (MTF) before down-sampling to the image-pixel grid. Acquisition noise is then added with a ramp noise-power spectrum (NPS), with subsequent smoothing through apodization filters. The features considered are lesion size, indistinct lesion boundary, and a nonuniform lesion interior. System resolution is implemented by an MTF with resolution (10% max.) of 0.47 or 0.58 cyc/mm. Apodization is implemented by a Shepp-Logan filter (Sinc profile) with various cutoffs. Six medically naïve subjects participated in the psychophysical studies, entailing training and testing components for each condition. Training consisted of staircase procedures to find the 80% correct threshold for each subject, and testing involved 2000 psychophysical trials at the threshold value for each subject. Human-observer performance is compared to the Ideal Observer to generate estimates of task efficiency. The significance of imaging factors is assessed using ANOVA. Classification images are used to estimate the linear template weights used by subjects to perform these tasks. Classification-image spectra are used to analyze subject weights in the spatial-frequency domain. Overall, average observer efficiency is relatively low in these experiments (10%-40%) relative to detection and localization studies reported previously. We find significant effects for feature type and apodization level on observer efficiency. Somewhat surprisingly, system resolution is not a significant factor. Efficiency effects of the different features appear to be well explained by the profile of the linear templates in the classification images. Increasingly strong apodization is found to both increase the classification-image weights and to increase the mean-frequency of the classification-image spectra. A secondary analysis of "Unapodized" classification images shows that this is largely due to observers undoing (inverting) the effects of apodization filters. These studies demonstrate that human observers can be relatively inefficient at feature-discrimination tasks in ramp-spectrum noise. Observers appear to be adapting to frequency suppression implemented in apodization filters, but there are residual effects that are not explained by spatial weighting patterns. The studies also suggest that the mechanisms for improving performance through the application of noise-control filters may require further investigation.

One-Stage Methods of Computer Vision Object Detection to Classify Carious Lesions from Smartphone Imaging

The current study aimed to implement and validate an automation system to detect carious lesions from smartphone images using different one-stage deep learning techniques. 233 images of carious lesions were captured using a smartphone camera system at 1432 × 1375 pixels, then classified and screened according to a visual caries classification index. Following data augmentation, the YOLO v5 model for object detection was used. After training the model with 1452 images at 640 × 588 pixel resolution, which included the ones that were created via image augmentation, a discrimination experiment was performed. Diagnostic indicators such as true positive, true negative, false positive, false negative, and mean average precision were used to analyze object detection performance and segmentation of systems. YOLO v5X and YOLO v5M models achieved superior performance over the other models on the same dataset. YOLO v5X’s mAP was 0.727, precision was 0.731, and recall was 0.729, which was higher than other models of YOLO v5, which generated 64% accuracy, with YOLO v5M producing slightly inferior results. Overall mAPs of 0.70, precision of 0.712, and recall of 0.708 were achieved. Object detection through the current YOLO models was able to successfully extract and classify regions of carious lesions from smartphone photographs of in vitro tooth specimens with reasonable accuracy. YOLO v5M was better fit to detect carious microcavitations while YOLO v5X was able to detect carious changes without cavitation. No single model was capable of adequately diagnosing all classifications of carious lesions.

Visual motion discrimination experiments reveal small differences between males and females

Recent results have shown that males have lower duration thresholds for motion direction discrimination than females. Measuring contrast thresholds, a previous study has shown that males have a greater sensitivity to fine details and fast flickering stimuli than females, and that females have a higher sensitivity to low spatial frequencies modulated at low temporal frequencies. Here, we present the data of a contrast-detection motion discrimination experiment and a reanalysis of four different motion discrimination experiments where we compare duration thresholds for males and females using different spatial frequencies, stimulus sizes, contrasts, and temporal frequencies (in two experiments, motion surround suppression was measured). Results from the main experiment and the reanalysis show that, in general, the association between sex and contrast and duration thresholds for motion discrimination is not significant, with males and females showing similar data patterns. Only the reanalysis of one out of four studies revealed different duration thresholds between males and females paired with a strong effect size supporting previous results in the literature, although motion surround suppression was identical between groups. Importantly, most of our results do not show significant differences between males and females in contrast and duration thresholds, suggesting that the sex variable may not be as relevant as previously claimed when testing visual motion discrimination.

A computational model to simulate spectral modulation and speech perception experiments of cochlear implant users.

Speech understanding in cochlear implant (CI) users presents large intersubject variability that may be related to different aspects of the peripheral auditory system, such as the electrode-nerve interface and neural health conditions. This variability makes it more challenging to proof differences in performance between different CI sound coding strategies in regular clinical studies, nevertheless, computational models can be helpful to assess the speech performance of CI users in an environment where all these physiological aspects can be controlled. In this study, differences in performance between three variants of the HiRes Fidelity 120 (F120) sound coding strategy are studied with a computational model. The computational model consists of (i) a processing stage with the sound coding strategy, (ii) a three-dimensional electrode-nerve interface that accounts for auditory nerve fiber (ANF) degeneration, (iii) a population of phenomenological ANF models, and (iv) a feature extractor algorithm to obtain the internal representation (IR) of the neural activity. As the back-end, the simulation framework for auditory discrimination experiments (FADE) was chosen. Two experiments relevant to speech understanding were performed: one related to spectral modulation threshold (SMT), and the other one related to speech reception threshold (SRT). These experiments included three different neural health conditions (healthy ANFs, and moderate and severe ANF degeneration). The F120 was configured to use sequential stimulation (F120-S), and simultaneous stimulation with two (F120-P) and three (F120-T) simultaneously active channels. Simultaneous stimulation causes electric interaction that smears the spectrotemporal information transmitted to the ANFs, and it has been hypothesized to lead to even worse information transmission in poor neural health conditions. In general, worse neural health conditions led to worse predicted performance; nevertheless, the detriment was small compared to clinical data. Results in SRT experiments indicated that performance with simultaneous stimulation, especially F120-T, were more affected by neural degeneration than with sequential stimulation. Results in SMT experiments showed no significant difference in performance. Although the proposed model in its current state is able to perform SMT and SRT experiments, it is not reliable to predict real CI users' performance yet. Nevertheless, improvements related to the ANF model, feature extraction, and predictor algorithm are discussed.

Learning from Freja the harbor porpoise to appreciate movements as an important behavioral readout to understand echolocation-based target discrimination

Toothed whales routinely discriminate and select prey via echolocation. Over a decade, the harbor porpoise Freja in Fjord and Bælt, Kerteminde, Denmark participated in a series of target discrimination experiments involving metal objects. Different from experiments in which the subjects were constrained to be stationary, Freja’s free and voluntary movements offered insights into the tightly coupled nature of acoustic sampling and movements in echolocation-based target discrimination. Compared with observations from wild animals with unknown prey, the controlled experimental task and stimuli made it possible to interpret her behavior based on predictable target echo features. Here, we test the hypothesis that an echolocating animal plans its movement based on prior echo returns to gather better target discrimination information. We trained Freja to approach and select a sphere against a spheroid in a two-alternative forced-choice discrimination task, and show that her performance and movement trajectory varied depending on the aspect ratio and presentation angle of the spheroid. Consistent with our hypothesis, we further show that Freja’s movement patterns were likely driven by echo information received earlier during the approach that is correlated with the discrimination difficulty. These results highlight movements as an important behavioral readout of the dynamic, closed-loop sensorimotor feedback in echolocation. [Work supported by ONR.]

Rotor Fault Diagnosis Method Using CNN-Based Transfer Learning with 2D Sound Spectrogram Analysis

This study discusses a failure detection algorithm that uses frequency analysis and artificial intelligence to determine whether a rotor used in an industrial setting has failed. A rotor is a standard component widely used in industrial sites, and continuous friction and corrosion frequently result in motor and bearing failures. As workers inspecting failure directly are at risk of serious accidents, an automated environment that can operate unmanned and a system for accurate failure determination are required. This study proposes an algorithm to detect faults by introducing convolutional neural networks (CNNs) after converting the fault sound from the rotor into a spectrogram through STFT analysis and visually processing it. A binary classifier for distinguishing between normal and failure states was added to the output part of the neural network structure used, which was based on the transfer learning methodology. We mounted the proposed structure on a designed embedded system to conduct performance discrimination experiments and analyze various outcome indicators using real-world fault data from various situations. The analysis revealed that failure could be detected in response to various normal and fault sounds of the field system and that both training and validation accuracy were greater than 99%. We further intend to investigate artificial intelligence algorithms that train and learn by classifying fault types into early, middle, and late stages to identify more specific faults.

Field-realistic acute exposure to glyphosate-based herbicide impairs fine-color discrimination in bumblebees.

Pollinator decline is a grave challenge worldwide. One of the main culprits for this decline is the widespread use of, and pollinators' chronic exposure to, agrochemicals. Here, we examined the effect of a field-realistic dose of the world's most commonly used pesticide, glyphosate-based herbicide (GBH), on bumblebee cognition. We experimentally tested bumblebee (Bombus terrestris) color and scent discrimination using acute GBH exposure, approximating a field-realistic dose from a day's foraging in a patch recently sprayed with GBH. In a 10-color discrimination experiment with five learning bouts, GBH treated bumblebees' learning rate fell to zero by third learning bout, whereas the control bees increased their performance in the last two bouts. In the memory test, the GBH treated bumblebees performed to near chance level, indicating that they had lost everything they had learned during the learning bouts, while the control bees were performing close to the level in their last learning bout. However, GBH did not affect bees' learning in a 2-color or 10-odor discrimination experiment, which suggests that the impact is limited to fine color learning and does not necessarily generalize to less specific tasks or other modalities. These results indicate that the widely used pesticide damages bumblebees' fine-color discrimination, which is essential to the pollinator's individual success and to colony fitness in complex foraging environments. Hence, our study suggests that acute sublethal exposure to GBH poses a greater threat to pollination-based ecosystem services than previously thought, and that tests for learning and memory should be integrated into pesticide risk assessment.

Biomechanically-Consistent Skin Stretch as an Intuitive Mechanism for Sensory Feedback: A Preliminary Investigation in the Lower Limb.

The proprioceptive loss accompanied by lower limb amputation can impair function and mobility. We explore a simple, mechanical skin-stretch array configured to generate superficial tissue behaviour that might occur with movement about an intact joint. Four adhesive pads attached around the circumference of the lower leg were connected via cords to a remote "foot" mounted on a ball joint attached to the underside of a fracture boot, such that "foot" reorientation would result in skin stretch. In two discrimination experiments performed with and without the connection, with no view of the mechanism, and with minimal training, unimpaired adults (i) estimated foot orientation following passive foot rotations (eight directions), either with or without contact between the lower leg and boot, and (ii) actively lowered the "foot" to estimate slope orientation (four directions). In (i), 56-60% of responses (depending on contact condition) were correct and 88-94% were either correct or one of the two adjacent choices. In (ii), 56% of responses were correct. In contrast, without the connection, participants performed near or no different to chance. A biomechanically-consistent skin stretch array may be an intuitive means to convey proprioceptive information from an artificial or poorly innervated joint.

A proof-of-concept meaning discrimination experiment to compile a word-in-context dataset for adjectives – A graph-based distributional approach

AbstractThe Word-in-Context corpus, which forms part of the SuperGLUE benchmark dataset, focuses on a specific sense disambiguation task: it has to be decided whether two occurrences of a given target word in two different contexts convey the same meaning or not. Unfortunately, the WiC database exhibits a relatively low consistency in terms of inter-annotator agreement, which implies that the meaning discrimination task is not well defined even for humans. The present paper aims at tackling this problem through anchoring semantic information to observable surface data. For doing so, we have experimented with a graph-based distributional approach, where both sparse and dense adjectival vector representations served as input. According to our expectations the algorithm is able to anchor the semantic information to contextual data, and therefore it is able to provide clear and explicit criteria as to when the same meaning should be assigned to the occurrences. Moreover, since this method does not rely on any external knowledge base, it should be suitable for any low- or medium-resourced language.