The effect of haptic guidance, aging, and initial skill level on motor learning of a steering task

In a previous study, we found that haptic guidance from a robotic steering wheel can improve short-term learning of steering of a simulated vehicle, in contrast to several studies of other tasks that had found that the guidance either impairs or does not aid motor learning. In this study, we examined whether haptic guidance-as-needed can improve long-term retention (across 1 week) of the steering task, with age and initial skill level as independent variables. Training with guidance-as-needed allowed all participants to learn to steer without experiencing large errors. For young participants (age 18–30), training with guidance-as-needed produced better long-term retention of driving skill than did training without guidance. For older participants (age 65–92), training with guidance-as-needed improved long-term retention in tracking error, but not significantly. However, for a subset of less skilled, older subjects, training with guidance-as-needed significantly improved long-term retention. The benefits of guidance-based training were most evident as an improved ability to straighten the vehicle direction when coming out of turns. In general, older participants not only systematically performed worse at the task than younger subjects (errors ～3 times greater), but also apparently learned more slowly, forgetting a greater percentage of the learned task during the 1 week layoffs between the experimental sessions. This study demonstrates that training with haptic guidance can benefit long-term retention of a driving skill for young and for some old drivers. Training with haptic guidance is more useful for people with less initial skill.     

Comparison of the haptic and visual deviations in a parallelity task

Deviations in both haptic and visual spatial experiments are thought to be caused by a biasing influence of an egocentric reference frame. The strength of this influence is strongly participant-dependent. By using a parallelity test, it is studied whether this strength is modality-independent. In both haptic and visual conditions, large, systematic and participant-dependent deviations were found. However, although the correlation between the haptic and visual deviations was significant, the explained variance due to a common factor was only 20%. Therefore, the degree to which a participant is “egocentric” depends on modality and possibly even more generally, on experimental condition.     

Motor learning with fading and growing haptic guidance

Haptic guidance has been shown to have both facilitatory and interfering effects on motor learning. Interfering effects have been hypothesized to result from the particular dynamic environment, which supports a passive role of the learner, and they should be attenuated by fading guidance. Facilitatory effects, in particular for dynamic movement characteristics, have been hypothesized to result from the high-quality information provided by haptic demonstration. If haptic demonstration provides particularly precise information about target movements, the motor system’s need for such information should more likely increase in the course of motor learning, in which case growing guidance should be more beneficial for learning. We contrasted fading and growing guidance in the course of learning a spatio-temporal motor pattern. To stimulate an active role of the learner, practice trials consisted of three phases, a visual demonstration of the target movement, a guided reproduction, and a reproduction without haptic guidance. Performance was assessed in terms of variable duration errors, relative-timing errors, variable path-length errors, and shape errors. Motor learning with growing and fading guidance turned out to be largely equivalent, so that the notion of an increasing optimal precision of haptic demonstrations, which matches a demand of increasingly precise information on the target movement, found no support. Duration errors declined only with fading, but not with growing guidance. Relative timing revealed a benefit of immediately preceding haptic demonstration, but learning was not different between the two practice protocols. This contrast between absolute and relative timing adds to other evidence according to which acquisition of these two aspects of motor timing involves different learning mechanisms. Whereas relative timing gained from immediately preceding haptic demonstration, but revealed no practice-related improvement in the presence of haptic guidance, the opposite pattern of results was found for the shape error. This finding is consistent with the claim that haptic demonstration is particularly efficient with respect to relative timing, but not with respect to spatial movement characteristics.     

The effect of category learning on visual attention and visual representation

Subordinate‐level category learning recruits neural resources associated with perceptual expertise, including the N250 component of the ERP, a posterolateral negative wave maximal between 230 and 330 ms. The N250 is a relatively late visual ERP and could plausibly be driven by attention to the features of categorized objects. Indeed, it has a latency and scalp distribution similar to the selection negativity (SN), an ERP component long known to be sensitive to attentional selection of target features. To clarify sensitivity of the N250 to attention and to more generally investigate the effect of category learning on attentional modulation of learned features, we independently manipulated subordinate‐level category learning and target detection in a speeded paradigm designed to optimally elicit the SN and accompanying frontal selection positivity (FSP). Participants first practiced categorizing a set of artificial animal stimuli and then performed a speeded target detection task on trained and untrained stimuli while ERPs were recorded. SN and FSP were roughly linearly related to the number of target features in the stimulus. Trained stimuli elicited a significantly larger N250 than untrained stimuli. The SN and N250 effects were additive, with all levels of target similarity equally affected by training, and had different time courses. Training had little effect on the FSP. The results suggest that (a) the N250 and SN have different sources, and (b) at the very least, the learning‐induced N250 indexes a different attentional subprocess from the target‐induced SN and could be driven by a different cognitive process altogether.     

Adaptation to visual feedback delays in a human manual tracking task

The time-course of human adaptation to spatial perturbations of visuomotor function (e.g. with prisms) is very short. However, it is not clear how rapid the adaptation to other aspects of perturbed feedback is. In this paper we report the adaptation to delayed visual feedback. Three groups of six subjects tracked unpredictable, continuously moving targets using a hand-held joystick while visual feedback of the joystick position was delayed (0 ms, 200 ms or 300 ms). Subjects clearly adapted to the delay, with a significant drop in tracking error, but changes in more subtle aspects of their tracking behaviour (such as changes in intermittency and their "impulse response functions") were not consistently observed. We suggest that the adaptation seen was consistent with the idea of there being a "delay component" in the internal processes used in manual tracking, as proposed in models such as the Smith predictor model.     

Delay improves performance on a haptic spatial matching task

Systematic deviations occur when blindfolded subjects set a test bar parallel to a reference bar in the horizontal plane using haptic information (Kappers and Koenderink 1999, Perception 28:781–795; Kappers 1999, Perception 28:1001–1012). These deviations are assumed to reflect the use of a combination of a biasing egocentric reference frame and an allocentric, more cognitive one (Kappers 2002, Acta Psychol 109:25–40). In two experiments, we have examined the effect of delay between the perception of a reference bar and the parallel setting of a test bar. In both experiments a 10-s delay improved performance. The improvement increased with a larger horizontal (left–right) distance between the bars. This improvement was interpreted as a shift from the egocentric towards the allocentric reference frame during the delay period. Electronic Publication     

Aging and electrocortical response to error feedback during a spatial learning task

Event-related potentials were collected as older and younger adults responded to error feedback in an adaptation of the Groton Maze Learning Test, an age-sensitive measure of spatial learning and executive skills expected to maximally involve anterior cingulate cortex (ACC). Older adults made more errors and produced smaller feedback-related negativities (FRNs) than young controls. LORETA source localization revealed that, for young adults, neural activation associated with the FRN was focused in ACC and was stronger to negative feedback. Older adults responded with less intense and less differentiated ACC activation, but FRN amplitudes did relate to error rate for the most difficult mazes. The feedback P3 was sensitive to negative feedback but played no role in the prediction of error for either group. These data reflect the selective age-related decline of ACC response but also its continued contribution to performance monitoring in aging.     

The effects of occipital ablation on a conditioned visual avoidance learning task in young and old rats

Previous data indicate that learning of a conditioned visual avoidance task following simultaneous complete bilateral occipital ablations is affected by the age of the organism as well as the postoperative recovery period. The present study investigated the performance of the avoidance task, following complete occipital ablations, in young and older rats given two different recovery intervals. Young rats given 10 days to recover performed more poorly than the young control animals; given 20 days to recover, young rats performed comparable to controls. Older rats given 10 days to recover performed significantly better than the controls; given 20 days to recover, older rats did not show a decrease in performance. The present finding is consistent with previous data which show that the occipital cortex exerted an inhibitory effect on the learning ability of the avoidance task in older rats, which, upon removal, enhanced the performance of these rats on the task.     

Contributions of delayed visual feedback and cognitive task load to postural dynamics

In this experiment, we examined the extent to which postural control is influenced by visual and cognitive task performance. Fourteen healthy young participants performed a balance task in eyes-open (EO) and delayed visual feedback (DVF) conditions. DVF was presented at delays ranging from 0 to 1200 ms in 300 ms increments. Cognitive load was implemented by a simple, serial arithmetic task. High and low-pass filtering (f_c = 0.3 Hz) distinguished LOW and HIGH frequency components, which were used to compute the variability of Anteroposterior (AP) Center of Pressure (COP) trajectories on fast (>0.3 Hz) and slow (<0.3 Hz) times cales. Imposed visual delay increased sway variability at both LOW and HIGH components. Cognitive task performance, however, influenced only the variability of fast (HIGH) sway components. Our results support distinct timescale mechanisms for postural control, but also demonstrate that vision predominantly influences low frequency components of postural sway. Moment-to-moment COP fluctuations are dependent on cognitive performance during delayed visual feedback postural control.     

The role of haptic feedback when manipulating nonrigid objects

Humans can learn to manipulate objects with complex dynamics, including nonrigid objects with internal degrees of freedom. The first aim of this study was to assess the contribution of haptic feedback when learning to manipulate a nonrigid object. The second aim was to evaluate how learning without haptic feedback influences subsequent learning with haptic feedback and vice versa. The task involved moving a simulated mass-attached to a grasped handle via a simulated, damped spring-to a target as quickly as possible. In the haptic plus vision (HV) condition, appropriate forces were applied to the handle, which was attached to a robot. In the vision only (V) condition, these forces were turned off. Participants completed 80 trials in each condition, with one-half starting with the HV condition. Both groups exhibited significant learning, as measured by movement time, in both conditions. For the condition performed first, initial performance, learning rate, and final performance were better with haptic feedback. Prior experience in the HV condition led to faster learning and better final performance in the V condition. However, prior experience in the V condition led to slower learning and worse final performance in the HV condition. In the V condition, all participants tended to keep the mass close to the hand. In the HV condition, participants who started with the HV condition allowed the mass to move away from the hand, whereas participants who started with the V condition continued to keep the mass close to the hand. We conclude that haptic feedback as well as prior experience with haptic feedback enhance the ability to control nonrigid objects and that training without haptic feedback can lead to persisting detrimental effects when subsequently dealing with haptic feedback.     

The effect of changed visual feedback on intention tremor in multiple sclerosis

In patients with multiple sclerosis (MS), intention tremor amplitude is enhanced during the visually guided compared to the memory guided motor tasks. In the present study, the effect of visual feedback on intention tremor was investigated during visually guided wrist step-tracking tasks. Specifically, visual feedback of the hand was provided either instantly or averaged over different time windows. Thirteen MS patients with intention tremor and 14 healthy controls performed the wrist step-tracking task, while the visual representation of the actual hand position was displayed instantly or averaged over time windows of 150, 250 and 350 ms. It has been found in the patient group that, in association with a decreased initial error and decreased tremor amplitude on the screen, the amplitude of the actual performed tremor also decreased when visual feedback was changed. The tremor reduction was not different between conditions with manipulated feedback, although delays in presenting visual feedback of the hand position increased when the time window was larger. The reduction in overall tremor amplitude was unlikely related to other factors, such as eye fixation deficits or the speed of the primary hand movement. These results suggest that hand tremor severity is dependent on the visual feedback of position and movement errors.     

Overcoming the guidance effect in motor skill learning: feedback all the time can be beneficial

Extensive research has shown that augmented feedback presented too often can create a dependency on the feedback and hinder long-term memory formation of a motor skill. This dependency has been labeled the guidance effect, and one way to overcome the guidance effect is to reduce how often augmented feedback is presented during training. In two experiments, participants were presented with visual augmented feedback during every trial in a 5-min training interval. Participants were provided visual augmented feedback in the form of a Lissajous template of a 1:2 multi-frequency pattern and a cursor representing the coordination between the limbs. Some participants were trained with the cursor superimposed (behind group) on the Lissajous template, and others were trained with the cursor presented in a separate window (side group) from the Lissajous template. In experiment 1, motion of the end-effectors was constrained to the medial–lateral direction in the horizontal plane. In experiment 2, end-effector motion was possible in both the medial–lateral and anterior–posterior directions in the horizontal plane. The location of the cursor did not influence performance during the 5-min training interval in either experiment. After a 15-min break, a retention test performed without the visual feedback provided by the cursor revealed that the behind groups’ performance was guided by the visual feedback in both experiments, whereas the side groups were able to perform without visual feedback. In experiment two, the side group’s performance without feedback was influenced when anterior–posterior motion was not constrained; however, the extent of the guidance effect was significantly less compared to the behind trained group in both experiments. The results show that the emergence of guided motor performance depends on the format of the display that provides visually based augmented feedback, and not just on how often the feedback is provided. In conclusion, visually based augmented feedback leads to the simultaneous development of a spatial and motor representation of the task. The behind format led to a dependence on the spatial representation developed during training, while the side format facilitated the development of the motor representation as a means to overcome guidance.     

The impact of light fingertip touch on haptic cortical processing during a standing balance task

Availability of fingertip touch onto a stable surface reduces body sway for subjects standing with eyes closed. This is largely associated with sensory feedback from the fingertip when mechanical load is limited. Here, it is possible that the central nervous system facilitates cortical sensory processing to augment feedback to control upright stance. To test this, we compared cortical sensory excitability between tasks with and without light finger touch while standing. Subjects stood in tandem on a force plate with eyes closed while lightly touching a stable surface with the index finger. This was, in two different studies, compared to: (1) no haptic contact or (2) light touch on an object not referenced to balance. Throughout testing, the median nerve was stimulated and electroencephalography was used to measure somatosensory evoked potentials (SEPs). As expected, availability of stable light touch reduced medial–lateral COP sway. Peak amplitudes for SEP components revealed reduced P100 (48%), but increased P50 (31%), N140 (80%), and P200 (20%) during stable touch versus no touch. The modulation of P50 and N140 was no longer present when comparing stable to control (touch), which suggested that attending to touch on either surface, regardless of stability reference, accounted for these changes. Conversely, P200 was increased (19%) when touching the stable surface. Our data show SEP modulation during a standing balance task related to hand contact. Facilitation of P200 in particular may indicate task-specific regulation of the cortical representation of fingertip afferent input when it is relevant to providing stable cues for static balance control.     

The visual guidance of catching

In order to explore the nature and amount of information in the optic array used by subjects required to carry out one-handed catching actions, the optical expansion pattern (using a deflating ball) and the duration of viewing time (using liquid crystal spectacles) of the ball were varied. Subjects were required to catch luminous balls (two of constant physical size and one of changing physical size during approach) attached to a pendulum in a totally dark room, while the liquid spectacles were closed at 0, 100, 200 or 300 ms before hand-ball contact. The results confirmed previous findings that the timing of the catching action is based on retinal expansion information and that conclusion was strengthened when an additional dependent variable (time of the maximal opening velocity of the grasp) was used. Further, for the viewing time duration manipulations, the time of the maximal closing velocity of the hand was later, while no effect was found on the time of the maximal opening velocity, when the last 300 ms of the trajectory of the ball was occluded. Adjustments to the catching action in response to the different ball sizes under the 0 ms condition differed significantly from the adjustments under the 300 ms condition. Both findings point to the importance of relative optical expansion information, available between 300 and 200 ms before ball-hand contact, in maintaining a (relatively) continuous perception-action coupling in the act of catching.     

Paradoxical sleep increases predict successful learning in a complex operant task

The sleep of 16 Sprague-Dawley rats was continuously monitored both before and during task acquisition. Eight rats were exposed in turn to autoshaping, 10-response fixed-ratio (FR 10), and sequential operant (SO) tasks. They all learned the autoshaping and FR 10 tasks with equal ease. However, 4 rats were unable to solve the SO task and were labeled as slow learning (SL). The other 4 rats did learn the task and were labeled as fast learning (FL). FL rats exhibited more paradoxical sleep than either the SL group or the two control groups. The results suggest a subtle, but as yet unknown, biological difference between FL and SL rats within the Sprague-Dawley strain.     

The effect of feature saliency on haptic subitizing

‘Subitizing’ refers to fast and error-free numerosity judgment for small (<4) sets of items. For larger sets, the slower process of ‘counting’ is used. Counting has a serial character, whereas subitizing is believed to have a parallel character. While subitizing was initially found in vision, it has been shown to exist in touch as well. In vision, it has been demonstrated that adding distractor items to a set of target items influences numerosity judgment of the target items. Subitizing was in this case only possible if the distractor item is highly salient among the targets. In the present study, we investigated the effect of adding a distractor item on haptic judgement of a set of target items. To this end, we asked subjects to judge the number of spheres grasped in their hand. Either a cube or an ellipsoid could be added to the set. A cube among spheres has been shown to be highly salient, while an ellipsoid among spheres is not. Our results show that adding a distractor item led to an increase in the response time slopes regardless of the distractor shape. Subitizing was, however, only possible in the case of a salient distractor. This is in agreement with results from vision.     

The effect of early visual experience on spatial maze learning in rats

In the 1st of 2 experiments on spatial ability, groups of sighted and blind, light-reared (LR) and dark-reared (DR) rats were tested on a series of (Hebb-Williams) maze problems and their reversals under appetitive and aversive reinforcement conditions. Significant effects due to early rearing conditions, vision at time of testing, and problem were found. Dark-reared rats learned the problems whose solution depended on nonvisual cues more slowly than LR animals. Blindness at time of testing had a significantly adverse effect on the performance of LR and DR rats on all problems, but a significantly greater effect in the DR animals. In a 2nd experiment DR rats were also found to perform less effectively than LR rats on a 17-arm radial maze throughout a 36-day period during which variations in the task were introduced. The results reveal the impact of early visual experience on the development of the ability to acquire spatial concepts.