0000000000004032
AUTHOR
Robert M. French
The subtlety of sameness: a theory and computer model of analogy-making
International audience; The research described in this book is based on the premise that human analogy-making is an extension of our constant background process of perceiving - in other words, that analogy-making and the perception of sameness are two sides of the same coin. At the heart of the author's theory and computer model of analogy-making is the idea that the building-up and the manipulation of representations are inseparable aspects of mental functioning, in contrast to traditional AI models of high-level cognitive processes, which have almost always depended on a clean separation. A computer program called Tabletop forms analogies in a microdomain consisting of everyday objects on…
The language of emotion in short blog texts
Emotion is central to human interactions, and automatic detection could enhance our experience with technologies. We investigate the linguistic expression of fine-grained emotion in 50 and 200 word samples of real blog texts previously coded by expert and naive raters. Content analysis (LIWC) reveals angry authors use more affective language and negative affect words, and that joyful authors use more positive affect words. Additionally, a co-occurrence semantic space approach (LSA) was able to identify fear (which naive human emotion raters could not do). We relate our findings to human emotion perception and note potential computational applications.
Five Ways in Which Computational Modeling Can Help Advance Cognitive Science
Abstract There is a rich tradition of building computational models in cognitive science, but modeling, theoretical, and experimental research are not as tightly integrated as they could be. In this paper, we show that computational techniques—even simple ones that are straightforward to use—can greatly facilitate designing, implementing, and analyzing experiments, and generally help lift research to a new level. We focus on the domain of artificial grammar learning, and we give five concrete examples in this domain for (a) formalizing and clarifying theories, (b) generating stimuli, (c) visualization, (d) model selection, and (e) exploring the hypothesis space.
Emotion rating from short blog texts
Being able to automatically perceive a variety of emotions from text alone has potentially important applications in CMC and HCI that range from identifying mood from online posts to enabling dynamically adaptive interfaces. However, such ability has not been proven in human raters or computational systems. Here we examine the ability of naive raters of emotion to detect one of eight emotional categories from 50 and 200 word samples of real blog text. Using expert raters as a 'gold standard', naive-expert rater agreement increased with longer texts, and was high for ratings of joy, disgust, anger and anticipation, but low for acceptance and 'neutral' texts. We discuss these findings in ligh…
The development of analogy making in children: cognitive load and executive functions.
The aim of the current study was to investigate the performance of 6-, 8-, and 14-year-olds on an analogy-making task involving analogies in which there are competing perceptual and relational matches. We hypothesized that the selection of the common relational structure requires the inhibition of other salient features, in particular, perceptual matches. Using an A:B::C:D paradigm, we showed that children’s performance in analogy-making tasks depends crucially on the nature of the distractors. Children chose more perceptual distractors having a common feature with C compared with A or B (Experiment 1). In addition, they were also influenced by unstructured random textures. When measuring r…
Implicit learning and consciousness: An empirical, philosophical, and computational consensus in the making
International audience; Can you learn without knowing it? This controversial and much debated question forms the basis of this collection of essays as the authors discuss whether the measurable changes in behaviour that result from learning can ever remain entirely unconscious. Three issues central to the topic of implicit learning are raised. Firstly, the extent to which learning can be unconscious, and therefore implicit, is considered. Secondly, theories are developed regarding the nature of knowledge acquired in implicit learning situations. Finally, the idea that there are two separable independent processing systems in the brain, for implicit and explicit learning, is considered.Impli…
Dusting Off the Turing Test
Hold up both hands and spread your fingers apart. Now put your palms together and fold your two middle fingers down till the knuckles on both fingers touch each other. While holding this position, one after the other, open and close each pair of opposing fingers by an inch or so. Notice anything? Of course you did. But could a computer without a body and without human experiences ever answer that question or a million others like it? And even if recent revolutionary advances in collecting, storing, retrieving, and analyzing data lead to such a computer, would this machine qualify as “intelligent”?
Relational priming is to analogy-making as one-ball juggling is to seven-ball juggling
Relational priming is argued to be a deeply inadequate model of analogy-making because of its intrinsic inability to do analogies where the base and target domains share no common attributes and the mapped relations are different. The authors rely on carefully handcrafted representations to allow their model to make a complex analogy, seemingly unaware of the debate on this issue 15 years ago. Finally, they incorrectly assume the existence of fixed, context-independent relations between objects. Although relational priming may indeed play some role in analogy-making, it is an enormous – and unjustified – stretch to say that it is “centrally implicated in analogical reasoning” (sect. 2, para…
Visual tracking combined with hand-tracking improves time perception of moving stimuli
A number of studies have shown that performing a secondary task while executing a time-judgment task impairs performance on the latter task. However, this turns out not to be the case for certain motor secondary tasks. We show that concomitant secondary motor tasks involving pointing, when performed during a time-judgment task, can actually improve our time-judgment abilities. We compared adult participants' performance in a time-of-movement paradigm with visual pursuit-only and with visual pursuit plus hand pursuit. Rather than interfering with their estimation of stimulus movement duration, the addition of hand pursuit significantly improved their judgment. In addition, we considered the …
GAMIT - A Fading-Gaussian Activation Model of Interval-Timing: Unifying Prospective and Retrospective Time Estimation
Two recent findings constitute a serious challenge for all existing models of interval timing. First, Hass and Hermann (2012) have shown that only variance-based processes will lead to the scalar growth of error that is characteristic of human time judgments. Secondly, a major meta-review of over one hundred studies of participants’ judgments of interval duration (Block et al., 2010) reveals a striking interaction between the way in which temporal judgments are queried (i.e., retrospectively or prospectively) and cognitive load. For retrospective time judgments, estimates under high cognitive load are longer than under low cognitive load. For prospective judgments, the reverse pattern holds…
Having your cake and eating it: Faster responses with reduced muscular activation while learning a temporal interval
International audience; We examined how motor responses to a stimulus evolve as individuals learn to predict when a stimulus will appear, by comparing responses to a regular versus irregular stimulus train. The study was conducted with two groups of adults — one responded to the regular appearance of a visual stimulus every 3 s (R group) and the second responded to the irregular presentation of the same stimulus (IR group) at intervals varying between 2 and 4 s. Participants responded to the appearance of the stimulus by bending over to press a button that was slightly out of reach. This whole body reach requires muscular activation at the ankles. Over the course of 50 consecutive responses…
Grandmother cells: much ado about nothing
International audience; We do not dispute the possibility of the existence in the brain of “grandmother cells”, which are very finely tuned neurons that fire only in the presence of specific objects or categories. However, we question the causal efficacy of such neurons at the functional or behaviour level. We claim that, even though very familiar items, such as “my grandmother”, may well have associated grandmother neurons, these neurons have very little, or no impact on the actual recognition of my grandmother. A study by Thomas, Van Hulle, and Vogels [(2002). Encoding of categories by noncategory-specific neurons in the inferior temporal cortex. Journal of Cognitive Neuroscience, 13, 190…
Visual Strategies in Analogical Reasoning Development: A New Method for Classifying Scanpaths
International audience; Development of analogical reasoning is often explained by general maturation of executive functions. A consequence of the involvement of executive functions would be that children and adults differ in the visual strategies they apply when solving analogical problems. Since visual strategies can be studied by means of eye-tracking, we compared the visual scanpaths of children and adults in three different analogical reasoning tasks. This comparison was done by means of a novel technique that combined a recently developed algorithm for computing a “distance” between any pair of scanpaths (Jarodzka, Holmqvist, & Nyström, 2010), multidimensional scaling (MDS), and a neur…
Interactive Effects of Explicit Emergent Structure: A Major Challenge for Cognitive Computational Modeling
International audience; David Marr's (1982) three-level analysis of computational cognition argues for three distinct levels of cognitive information processingnamely, the computational, representational, and implementational levels. But Marr's levels areand were meant to bedescriptive, rather than interactive and dynamic. For this reason, we suggest that, had Marr been writing today, he might well have gone even farther in his analysis, including the emergence of structurein particular, explicit structure at the conceptual levelfrom lower levels, and the effect of explicit emergent structures on the level (or levels) that gave rise to them. The message is that today's cognitive scientists …
An unsupervised dual-network connectionist model of rule emergence in category learning
We develop an unsupervised dual-network connectionist model of category learning in which rules gradually emerge from a standard Kohonen network. The architecture is based on the interaction of a statistical-learning (Kohonen) network and a competitive-learning rule network. The rules that emerge in the rule network are weightings of individual features according to their importance for categorisation. Once the combined system has learned a particular rule, it de-emphasizes those features that are not sufficient for categorisation, thus allowing correct classification of novel, but atypical, stimuli, for which a standard Kohonen network fails. We explain the principles and architectural det…
Embodiment and the origin of interval timing : kinematic and electromyographic data
International audience; Recent evidence suggests that interval timing (the judgment of durations lasting from approximately 500 ms. to a few minutes) is closely coupled to the action control system. We used surface electromyography (EMG) and motion capture technology to explore the emergence of this coupling in 4-, 6-, and 8-month-olds. We engaged infants in an active and socially relevant arm-raising task with seven cycles and response period. In one condition, cycles were slow (every 4 s); in another, they were fast (every 2 s). In the slow condition, we found evidence of time-locked sub-threshold EMG activity even in the absence of any observed overt motor responses at all three ages. Th…
Explaining children’s failure in analogy making tasks: A problem of focus of attention?
International audience; Analogical reasoning is commonly recognized as essential to human cognition, but young children often perform poorly in the classical A:B::C:? analogical reasoning task. Previous eye-tracking results have shown that children did not visually explore the A:B pair as much as adults in this task. We hypothesized that this lack of exploration could help account for the low scores of children in comparison to adults. The present study shows that children’ performance improves significantly if they are required to look at and process the A:B pair before they are shown the full A:B::C:? problem. This confirms our hypothesis that the A:B pair is insufficiently processed by c…
Pupil Diameter May Reflect Motor Control and Learning
International audience; Non–luminance-mediated changes in pupil diameter have been used since the first studies by Darwin in 1872 as indicators of clinical, cognitive, and arousal states. However, the relation between processes involved in motor control and changes in pupil diameter remains largely unknown. Twenty participants attempted to compensate random walks of a cursor with a computer mouse to restrain its trajectory within a target circle while the authors recorded their pupil diameters. Two conditions allowed the authors to experimentally manipulate the motor and cognitive components of the task. First, the step size of the cursor's random walk was either large or small leading to 2…
The Application of Machine Learning Algorithms to the Analysis of Electromyographic Patterns From Arthritic Patients
The main aim of our study was to investigate the possibility of applying machine learning techniques to the analysis of electromyographic patterns (EMG) collected from arthritic patients during gait. The EMG recordings were collected from the lower limbs of patients with arthritis and compared with those of healthy subjects (CO) with no musculoskeletal disorder. The study involved subjects suffering from two forms of arthritis, viz, rheumatoid arthritis (RA) and hip osteoarthritis (OA). The analysis of the data was plagued by two problems which frequently render the analysis of this type of data extremely difficult. One was the small number of human subjects that could be included in the in…
A classification study of kinematic gait trajectories in hip osteoarthritis
The clinical evaluation of patients in hip osteoarthritis is often done using patient questionnaires. While this provides important information it is also necessary to continue developing objective measures. In this work we further investigate the studies concerning the use of 3D gait analysis to attain this goal. The gait analysis was associated with machine learning methods in order to provide a direct measure of patient control gait discrimination. The applied machine learning method was the support vector machine (SVM). Applying the SVM on all the measured kinematic trajectories, we were able to classify individual patient and control gait cycles with a mean success rate of 88%. With th…
A new manifesto for child development research
AbstractThis book is an excellent manifesto for future work in child development. It presents a multidisciplinary approach that clearly demonstrates the value of integrating modeling, neuroscience, and behavior to explore the mechanisms underlying development and to show how internal context-dependent representations arise and are modified during development. Its only major flaw is to have given short shrift to the study of the role of genetics on development.
Shifts in Key Time Points and Strategies for a Multisegment Motor Task in Healthy Aging Subjects
International audience; In this study, we compared key temporal points in the whole body pointing movement of healthy aging and young subjects. During this movement, subject leans forward from a standing position to reach a target. As it involves forward inclination of the trunk, the movement creates a risk for falling. We examined two strategic time points during the task-first, the crossover point where the velocity of the center of mass (CoM) in the vertical dimension outstripped the velocity in the anteroposterior dimension and secondly, the time to peak of the CoM velocity profile. Transitions to stabilizing postures occur at these time points. They both occurred earlier in aging subje…
Gödel, Escher, Bach : Les brins d'une guirlande éternelle
International audience; Quel rapport y a-t-il entre la musique de Jean-Sébastien Bach, les dessins du graveur néerlandais Maurits Escher, et le célèbre théorème du logicien autrichien Kurt Gödel ? Du premier, on connaît des pièces lisibles indifféremment dans les deux sens, ou répétant le même motif sous des formes toujours nouvelles ; Escher, lui, nous a laissé des images paradoxales de fontaines s'alimentant elles-mêmes, de bandes de Möbius infinies ou de mains s'autodessinant. De Gödel enfin, vient cet étrange théorème posant une limite à la capacité des mathématiques à démontrer leurs propres théorèmes. "Autoréférence" est ainsi le maître mot d'un récit fleuve, devenu livre-culte, d'une…
Turn, turn, turn : perceiving global and local, clockwise and counterclockwise rotations
International audience; We developed a series of dynamic geometric Navon figures in order to study global/local rotation processing. These figures consist of a global figure (a triangle or a square) made up of local figures (also triangles or squares). Both global and local figures can rotate in either clockwise or counterclockwise directions independently. We found that there is no right or left visual field perceptual advantage for either the global or local levels of these figures, as in Sergent (1982). We did, however, find a significant processing advantage for clockwise motion compared to counterclockwise motion. We also found a highly significant interaction between the detection of …
Catégorisation asymétrique de séquences de hauteurs musicales
Asymmetric Categorization in the Sequential Auditory Domain An unusual visual category learning asymmetry in infants was observed by Quinn, Eimas, and Rosenkrantz (1993). A series of experiments and simulations seemed to show that this asymmetry was due the perceptual inclusion of the cat category within the dog category because of the greater perceptual variability of the distributions of the visual features of dogs compared to cats (Mareschal & French, 1997; Mareschal, French, & Quinn, 2000; French, Mermillod, Quinn, & Mareschal, 2001; French, Mareschal, Mermillod, & Quinn, 2004). In the present paper, we explore whether this asymmetric categorization phenomenon generalizes to the auditor…
Moving beyond the Turing test
Computers interacting with, not imitating, humans is the way forward.
KAMA: A Temperature-Driven Model of Mate Choice Using Dynamic Partner Representations
KAMA is a model of mate-choice based on a gradual, stochastic process of building up representations of potential partners through encounters and dating, ultimately leading to marriage. Individuals must attempt to find a suitable mate in a limited amount of time with only partial knowledge of the individuals in the pool of potential candidates. Individuals have multiple-valued character profiles, which describe a number of their characteristics (physical beauty, potential earning power, etc.), as well as preference profiles, that specify their degree of preference for those characteristics in members of the opposite sex. A process of encounters and dating allows individuals to gradually bui…
Unifying prospective and retrospective interval-time estimation: A fading-Gaussian activation-based model of interval-timing
International audience; Hass and Hermann (2012) have shown that only variance-based processes will lead to the scalar growth of error that is characteristic of human time judgments. Secondly, a major meta-review of over one hundred studies (Block et al., 2010) reveals a striking interaction between the way in which temporal judgments are queried and cognitive load on participants’ judgments of interval duration. For retrospective time judgments, estimates under high cognitive load are longer than under low cognitive load. For prospective judgments, the reverse pattern holds, with increased cognitive load leading to shorter estimates. We describe GAMIT, a Gaussian spreading-activation model,…
Computational modeling in cognitive science: a manifesto for change.
Computational modeling has long been one of the traditional pillars of cognitive science. Unfortunately, the computer models of cognition being developed today have not kept up with the enormous changes that have taken place in computer technology and, especially, in human-computer interfaces. For all intents and purposes, modeling is still done today as it was 25, or even 35, years ago. Everyone still programs in his or her own favorite programming language, source code is rarely made available, accessibility of models to non-programming researchers is essentially non-existent, and even for other modelers, the profusion of source code in a multitude of programming languages, written witho…
Learning to perceive time: A connectionist, memory-decay model of the development of interval timing in infants
International audience; We present the first developmental model of interval timing. It is a memory-based connectionist model of how infants learn to perceive time. It has two novel features that are not found in other models. First, it uses the uncertainty of a memory for an event as an index of how long ago that event happened. Secondly, embodiment – specifically, infant motor activity – is crucial to the calibration of time-perception both within and across sensory modalities. We describe the model and present three simulations which show (1) how it uses sensory memory uncertainty and bodily representaions to index time, (2) that the scalar property of interval timing (Gibbon, 1977) emer…
The red tooth hypothesis: A computational model of predator-prey relations, protean escape behavior and sexual reproduction
This paper presents an extension of the Red Queen Hypothesis (hereafter, RQH) that we call the Red Tooth Hypothesis (RTH). This hypothesis suggests that predator-prey relations may play a role in the maintenance of sexual reproduction in many higher animals. RTH is based on an interaction between learning on the part of predators and evolution on the part of prey. We present a simple predator-prey computer simulation that illustrates the effects of this interaction. This simulation suggests that the optimal escape strategy from the prey's standpoint would be to have a small number of highly reflexive, largely innate (and, therefore, very fast) escape patterns, but that would also be unlearn…
Across space and time : infants learn from backward and forward visual statistics
International audience; This study investigates whether infants are sensitive to backward and forward transitional probabilities within temporal and spatial visual streams. Two groups of 8-month-old infants were familiarized with an artificial grammar of shapes, comprising backward and forward base pairs (i.e. two shapes linked by strong backward or forward transitional probability) and part-pairs (i.e. two shapes with weak transitional probabilities in both directions). One group viewed the continuous visual stream as a temporal sequence, while the other group viewed the same stream as a spatial array. Following familiarization, infants looked longer at test trials containing part-pairs th…
From Associations to Rules: Connectionist Models of Behavior and Cognition
International audience; This book introduces a host of connectionist models of cognition and behavior. The major areas covered are high-level cognition, language, categorization and visual perception, and sensory and attentional processing. All of the articles cover unpublished research work. The key contribution of this book is that it focuses exclusively on the advances in connectionist modeling in psychology. The papers are relatively short, and were explicitly written to be accessible to both connectionist modelers and experimental psychologists.
Missing the Forest for the Trees: Why Cognitive Science Circa 2019 Is Alive and Well
International audience; Núñez and colleagues (2019) chronicle in extraordinary detail the "demise" of cognitive science, as it was first defined in the late 1970s. The problem is that their account, however accurate, misses the forest for the trees. Cognitive science circa 2019 is alive and well; it just has not followed the path anticipated by its founders over 40 years ago.
TRACX2: a RAAM -like autoencoder modeling graded chunking in infant visual -sequence learning
International audience; Even newborn infants are able to extract structure from a stream of sensory inputs and yet, how this is achieved remains largely a mystery. We present a connectionist autoencoder model, TRACX2, that learns to extract sequence structure by gradually constructing chunks, storing these chunks in a distributed manner across its synaptic weights, and recognizing these chunks when they re-occur in the input stream. Chunks are graded rather than all-or-none in nature and during learning their component parts become ever more tightly bound together. TRACX2 successfully models data from four experiments from the infant visual statistical-learning literature, including tasks i…