Decision-making in robotics and psychology: A distributed account

Decision-making is usually a secondary topic in psychology, relegated to the last chapters of textbooks. The psychological study of decision-making assumes a certain conception of its nature and mechanisms that has been shown wrong by research in robotics. Robotics indicate that decision-making is not—or at least not only—an intellectual task, but also a process of dynamic behavioral control, mediated by embodied and situated sensorimotor interaction. The implications of this conception for psychology are discussed.     

Linking impulsivity and inhibitory control using manual and oculomotor response inhibition tasks

Separate cognitive processes govern the inhibitory control of manual and oculomotor movements. Despite this fundamental distinction, little is known about how these inhibitory control processes relate to more complex domains of behavioral functioning. This study sought to determine how these inhibitory control mechanisms relate to broadly defined domains of impulsive behavior. Thirty adults with attention-deficit/hyperactivity disorder (ADHD) and 28 comparison adults performed behavioral measures of inhibitory control and completed impulsivity inventories. Results suggest that oculomotor inhibitory control, but not manual inhibitory control, is related to specific domains of self-reported impulsivity. This finding was limited to the ADHD group; no significant relations between inhibitory control and impulsivity were found in comparison adults. These results highlight the heterogeneity of inhibitory control processes and their differential relations to different facets of impulsivity.     

Temporal-range estimation of multiple objects: evidence for an early bottleneck

When making parallel time-to-contact (TTC) estimates of two approaching objects, the two respective TTC estimates interfere with one another in an asymmetric fashion. The TTC of the later-arriving object is systematically overestimated, while the estimated TTC for the first-arriving object is as accurate as in a condition presenting only a single object. This asymmetric interference points to a processing bottleneck that could be due to early (e.g., during the estimation of the TTC from the optic flow) or late (e.g., during the timing of the response or the motor execution) constraints in the TTC estimation process. We used a Sperling-like prediction-motion task to differentiate between these two possibilities. Participants produced an absolute estimate of the TTC of only one of two objects approaching a target line. The target object to which the response was to be made was indicated by an auditory cue that occurred either at motion-onset or at the instant at which the two objects disappeared from the screen (occlusion-onset). The cue at motion-onset should disengage visual processing of the irrelevant stimulus. The cue at occlusion-onset, in contrast, requires visual processing of both relevant and irrelevant stimulus until occlusion. A single-object condition was introduced as a control condition. Results show symmetric interference in the motion-onset condition. In the occlusion-onset condition however, the results were congruent with asymmetric interference. Thus, the processing bottleneck in TTC estimation is originating at the earlier stages.     

Grasping of extrafoveal targets: A robotic model

We present a computational model of grasping of non-fixated (extrafoveal) target objects which is implemented on a robot setup, consisting of a robot arm with cameras and gripper. This model is based on the premotor theory of attention (Rizzolatti et al., 1994) which states that spatial attention is a consequence of the preparation of goal-directed, spatially coded movements (especially saccadic eye movements). In our model, we add the hypothesis that saccade planning is accompanied by the prediction of the retinal images after the saccade. The foveal region of these predicted images can be used to determine the orientation and shape of objects at the target location of the attention shift. This information is necessary for precise grasping. Our model consists of a saccade controller for target fixation, a visual forward model for the prediction of retinal images, and an arm controller which generates arm postures for grasping. We compare the precision of the robotic model in different task conditions, among them grasping (1) towards fixated target objects using the actual retinal images, (2) towards non-fixated target objects using visual prediction, and (3) towards non-fixated target objects without visual prediction. The first and second setting result in good grasping performance, while the third setting causes considerable errors of the gripper orientation, demonstrating that visual prediction might be an important component of eye–hand coordination. Finally, based on the present study we argue that the use of robots is a valuable research methodology within psychology.     

Neuro-cognitive mechanisms of decision making in joint action: a human-robot interaction study

In this paper we present a model for action preparation and decision making in cooperative tasks that is inspired by recent experimental findings about the neuro-cognitive mechanisms supporting joint action in humans. It implements the coordination of actions and goals among the partners as a dynamic process that integrates contextual cues, shared task knowledge and predicted outcome of others’ motor behavior. The control architecture is formalized by a system of coupled dynamic neural fields representing a distributed network of local but connected neural populations. Different pools of neurons encode task-relevant information about action means, task goals and context in the form of self-sustained activation patterns. These patterns are triggered by input from connected populations and evolve continuously in time under the influence of recurrent interactions. The dynamic model of joint action is evaluated in a task in which a robot and a human jointly construct a toy object. We show that the highly context sensitive mapping from action observation onto appropriate complementary actions allows coping with dynamically changing joint action situations.     

Modulating Fitts's Law: perceiving targets at the last placeholder

Fitts's Law predicts increasing movement times (MTs) with increasing movement amplitudes; however, when targets are placed in a structured perceptual array containing placeholders, MTs to targets in the last position are shorter than predicted. We conducted three experiments to determine if this modulation has a perceptual cause. Experiment 1, which used extremely diminished (three pixel) placeholders, showed that the modulation is not due to perceptual interference from neighboring placeholders. Experiment 2, which measured reaction times using a target detection task, showed that the modulation does not result from speeded perceptual processing at the last position of the array. Experiment 3, which measured accuracy using a masked letter-discrimination task, showed that the modulation does not result from the increased quality of perceptual representation at the last position of the array. Overall, these findings suggest that the changes in effectiveness of visual processing (less interference, speeded processing, and increased quality) at the last position in the perceptual array do not drive the modulation. Thus, while the locus of the Fitts's Law modulation appears to be in the movement planning stage, it is likely not due to perceptual mechanisms.     

Global precedence effect in audition and vision: evidence for similar cognitive styles across modalities

This study aimed to provide evidence for a Global Precedence Effect (GPE) in both vision and audition modalities. In order to parallel Navon's paradigm, a novel auditory task was designed in which hierarchical auditory stimuli were used to involve local and global processing. Participants were asked to process auditory and visual hierarchical patterns at the local or global level. In both modalities, a global-over-local advantage and a global interference on local processing were found. The other compelling result is a significant correlation between these effects across modalities. Evidence that the same participants exhibit similar processing style across modalities strongly supports the idea of a cognitive style to process information and common processing principle in perception.     

Value systems for developmental cognitive robotics: A survey

This paper surveys value systems for developmental cognitive robotics. A value system permits a biological brain to increase the likelihood of neural responses to selected external phenomena. Many machine learning algorithms capture the essence of this learning process. However, computational value systems aim not only to support learning, but also autonomous attention focus to direct learning. This combination of unsupervised attention focus and learning aims to address the grand challenge of autonomous mental development for machines. This survey examines existing value systems for developmental cognitive robotics in this context. We examine the definitions of value used—including recent pioneering work in intrinsic motivation as value—as well as initialisation strategies for innate values, update strategies for acquired value and the data structures used for storing value. We examine the extent to which existing value systems support attention focus, learning and prediction in an unsupervised setting. The types of robots and applications in which these value systems are used are also examined, as well as the ways that these applications are evaluated. Finally, we study the strengths and limitations of current value systems for developmental cognitive robots and conclude with a set of research challenges for this field.     

Exploring cross-task compatibility in perceiving and producing facial expressions using electromyography

Using a dual-task methodology we examined the interaction of perceiving and producing facial expressions. In one task, participants were asked to produce a smile or a frown (Task 2) in response to a tone stimulus. This auditory-facial task was embedded in a dual-task context, where the other task (Task 1) required a manual response to visual face stimuli (visual-manual task). These face stimuli showed facial expressions that were either compatible or incompatible to the to-be-produced facial expression. Both reaction times and error rates (measured by facial electromyography) revealed a robust stimulus–response compatibility effect across tasks, suggesting that perceived social actions automatically activate corresponding actions even if perceived and produced actions belong to different tasks. The dual-task nature of this compatibility effect further testifies that encoding of facial expressions is highly automatic.     

Computational modeling/cognitive robotics complements functional modeling/experimental psychology

This position paper explores the possible contributions to the science of psychology from insights obtained by building and experimenting with cognitive robots. First, the functional modeling characteristic of experimental psychology is discussed. Second, the computational modeling required for cognitive robotics is described, and possible experiments with them are illustrated. Next, we argue that cognitive developmental robots, robots that “live” through a development phase where they learn about their environments in several different modes, can provide additional benefits to the science of psychology. Finally, the reciprocal interactions between computational modeling/cognitive robotics and functional modeling/experimental psychology are explored. We conclude that each can contribute significantly to the other.     

Validation of a full body 3-D dynamic linked segment model

In studies related to human movement, linked segment models (LSM's) are often used to quantify forces and torques, generated in body joints. Some LSM's represent only a few body segments. Others, for instance used in studies on the control of whole body movements, include all body segments. As a consequence of the complexity of 3-dimensional (3-D) analyses, most LSM's are restricted to one plane of motion. However, in asymmetric movements this may result in a loss of relevant information. The aim of the current study was to develop and validate a 3-D LSM including all body segments. Braces with markers, attached to all body segments, were used to record the body movements. The validation of the model was accomplished by comparing the measured with the estimated ground reaction force and by comparing the torques at the lumbo-sacral joint that resulted from a bottom-up and a top-down mechanical analysis. For both comparisons, reasonable to good agreement was found. Sources of error that could not be analysed this way, were subjected to an additional sensitivity analysis. It was concluded that the internal validity of the current model is quite satisfactory.     

An investigation of the determinants of motor contagion in preschool children

The influence of action perception on action execution has been demonstrated by studies of motor contagion in which the observation of an action interferes with the concurrent execution of a different action. The current study extends prior work on the extent of motor contagion in early childhood, a period of development when the effects of action observation on action execution may be particularly salient. During a classroom story reading, children (mean age 4.8 years) were familiarized with two different-colored bears, one of which was used as a seemingly animate hand puppet while the other bear remained lifeless and inanimate. Children then completed a task in which they were instructed to move a stylus on a graphics tablet in the presence of background videos of each bear making horizontal arm movements which had biological (human-moved) or non-biological (machine-moved) origins. Motor contagion was assessed as the variability of stylus movements in the horizontal axis when children were instructed to produce vertical stylus movements. Significant levels of motor contagion were seen when children observed the previously animate bear in the non-biological motion condition and when they observed the previously inanimate bear in the biological motion condition. For future studies of social perception, this finding points to the potential importance of examining mismatches between prior experience with (or knowledge about) a particular agent and the subsequent behavior of that agent in a different context.     

The Simon effect in cognitive electrophysiology: A short review

In the Simon task, stimuli are presented laterally and typically a non-spatial stimulus dimension demands a lateralized choice response. Responses are faster when the task-irrelevant stimulus location and the response location correspond than when they do not correspond. The present paper explores the impact of the Simon task on cognitive electrophysiological research as well as the insights gained from event-related brain potentials (ERPs) in the attempt to uncover the hidden mechanisms underlying the Simon effect.     

Explicit and motoric dependent measures of geographical slant are dissociable: A reassessment of the findings of Durgin, Hajnal, Li, Tonge, and Stigliani (2010)

Durgin et al. (2010) argued that the apparent accuracy of the palmboard measure of geographical slant is accidental and reflects limitations in wrist flexion that reduce palmboard adjustments by just the right amount given the perceptual overestimations upon which they are based. This account is inconsistent with findings that verbal reports and palmboard adjustments are dissociable. In addition to previous evidence found for such dissociation, Durgin et al. also found verbal/palmboard dissociations in Experiment 2. Experiments 1 and 3 of Durgin et al. lacked verbal reports and instead compared palmboard adjustments to free-hand estimates in the context of small wooden surfaces. These experiments are not relevant to the issue of verbal/palmboard dissociability. Across studies, the accuracy of Durgin et al.'s palmboard implementation is far less than that found by others (Feresin & Agostini, 2007). The design of Durgin et al.'s Experiment 5 misrepresented the experimental conditions of Creem and Proffitt (1998), and consequently, the findings of this study have no bearing on the issue at hand.     

Structural model of employee involvement in skill development activity: The role of individual differences

We extend prior research on involvement in employee development activity by including prominent individual difference constructs that have been previously ignored in this area of research. These include two important personality characteristics (conscientiousness and openness to experience), mental ability and goal orientation constructs. We tested both mediated and direct effects of the variables. The sequence of relationships observed in the model was: general personality traits → development domain individual and situational variables → development domain motivational variables → involvement in development. Personality had indirect (mediated) effects on motivational and involvement constructs while goal orientation, as a development domain individual variable, had direct effects on motivational constructs. Mental ability had no effects. The study also replicated core relationships from prior research, but did so using an Internet sample and response medium. The model presented provides the most complete picture of behavior in this area to date. Important implications for advancement of theory, research and practice in the area of employee development behavior are discussed.     

Reinforcement of inhibitory positional priming by spatial working memory contents

This study examined whether positional priming of pop-out is modulated by spatial working memory (WM). In Experiments 1A and 1B, observers were presented with a to-be-memorized item (Experiment 1A: triangle; Experiment 1B: circle) prior to performing a block of search trials, with triangular stimulus configurations (one target, two distractors) on an underlying elliptical frame. Each search block was followed by a memory test (i.e., memory conditions). In Experiment 1C, the presentation of the memory item (triangle) was immediately followed by the memory test, and the search block was performed after this test (i.e., no-memory condition). It was found that targets presented at a previous target were detected faster and targets presented at a previous distractor location slower, relative to a previous empty (‘neutral’) stimulus location, respectively. While facilitatory position priming was uninfluenced by the memory manipulation, inhibitory positional priming was larger in the memory than no-memory condition, but only when the memory stimulus (triangle) matched the (triangle) configuration formed by the three search items. This finding of larger inhibitory priming under memory conditions provides further evidence that positional priming, rather than being purely bottom-up determined, is open to top-down influences.     

Emotional reactivity and force control: the influence of behavioral inhibition

Individual difference measures have been shown to alter emotional arousal and emotional arousal alters force production during force control tasks. In the current study we examined whether individual differences in behavioral inhibition influence force control during emotional image viewing. Subjects who scored high and low in behavioral inhibition (BIS) produced force with visual feedback for 5 s. Feedback was then removed and replaced by a mutilation, attack, erotica, or neutral image for 6 s. The magnitude and direction of error in force production during image presentation was compared between groups and across image type. The high BIS group displayed a relative increase in force production during exposure to attack and mutilation images compared to the low BIS group. Bias scores (i.e., comparison of unpleasant image to neutral or pleasant image) further confirmed these findings by demonstrating a relative increase in force for the high BIS group during attack and mutilation images as compared to erotica images, whereas the low BIS group displayed the reverse effect. Together these findings extend the premise of action readiness to demonstrate that dispositional differences in behavioral inhibition interact with emotional state to alter force production.     

Non-differential return of fear in humans after a reinstatement procedure

The present experiment investigated reinstatement of fear in humans using a differential fear conditioning preparation. In this experiment, one neutral stimulus (conditioned stimulus; CS+) was paired with an aversive stimulus (unconditioned stimulus; US) during the acquisition phase, while another neutral stimulus was not (CS−). This procedure led to a difference in responding between the CS+ and the CS− (i.e., differential conditioning). After this acquisition phase, an extinction phase followed, during which both CSs were presented without the US, resulting in a decrease in differential conditioned responding. Reinstatement refers to the return of extinguished conditioned responses due to the experience of US-only trials after the extinction phase. This phenomenon was investigated by presenting half of the participants (reinstatement group) with unpredictable USs after the extinction phase. The control group did not receive these USs after the extinction procedure. The results show that return of fear was clearly apparent after the reinstating USs. This return of fear was, however, not limited to the CS+. An increase in ‘conditioned’ responding was also observed for the control stimulus. This interesting observation will be discussed against the background of a number of recent theoretical conceptualizations of reinstatement.     

Coordination of complex bimanual multijoint movements under increasing cycling frequencies: The prevalence of mirror-image and translational symmetry

The present study examined the principles underlying inter and intralimb coordination constraints during performance of bimanual elbow–wrist movements at different cycling frequencies (from 0.75 Hz to 2.50 Hz). Participants performed eight coordination tasks that consisted of a combination of in-phase (IN) and/or anti-phase (AN) coordination modes between both elbows and wrists (interlimb), with isodirectional (Iso) or non-isodirectional (NonI) coordination modes within each limb (intralimb). As expected, the principle of muscle homology (in-phase coordination), giving rise to mirror symmetrical movements with respect to the mid-sagittal plane, had a powerful influence on the quality of global coordinative behavior both between and within limbs. When this principle was violated (i.e., when the anti-phase mode was introduced in one or both joint pairs), the non-isodirectional intralimb mode exhibited a (de)stabilizing role in coordination, which became more pronounced at higher cycling frequencies. However, pattern loss with increasing cycling frequency resulted not only in convergence toward the more stable in-phase patterns with the elbows and wrists but also to the anti-phase patterns (which were associated with directional compatibility of within-limb motions). Moreover, participants generally preserved their initial mode of coordination (either in-phase or anti-phase) in the proximal joints (i.e., elbows) while shifting from anti-phase to in-phase (or vice versa) with their distal joint pair (i.e., wrists). Taken together, these findings reflect the impact of two immanent types of symmetry in bimanual coordination: mirror-image and translational symmetry.