Andrea H. Mason

Effects of Visual Similarity on Serial Report and Item Recognition

Four experiments examined the sensitivity of visual short-term memory to visual pattern similarity. Experiment 1 showed that immediate serial memory for novel visual patterns was sensitive to similarity. Using an item reocgnition task, Experiment 2 showed that subjects learned the descriptions of sets of similar and dissimilar patterns at the same rate. But repeated presentations of patterns in a serial memory task again showed a marked and persistent similarity effect (Experiment 3). The final experiment showed the visual similarity effect in serial memory for patterns that had been previously learned. The results show that (a) serial memory for patterns is sensitive to visual similarity, (b) the visual similarity effect is not due to perceptual confusions but originates in memory, (c) there is a clear dissociation between item and order errors, and (d) the visual similarity effect survives articulatory suppression. Visual serial order memory and verbal serial recall appear to share several of these prop...

Target viewing time and velocity effects on prehension

Abstract The goal of the present study was to understand which characteristics (movement time or velocity) of target motion are important in the control and coordination of the transport and grasp-preshape components of prehensile movements during an interception task. Subjects were required to reach toward, grasp and lift an object as it entered a target area. Targets approached along a track at four velocities (500, 750, 1000 and 1250 mm/s) which were presented in two conditions. In the distance-controlled condition, targets moving at all velocities traveled the same distance. In the viewing-time-controlled condition, combinations of velocity and starting distances were performed such that the moving target was visible for 1000 ms for all trials. Analyses of kinematic data revealed that when, target distance was controlled, velocity affected all transport-dependent measures; however, when viewing time was controlled, these dependent measures were no longer affected by target velocity. Thus, the use of velocity information was limited in the viewing-time-controlled condition, and subjects used other information, such as target movement time, when generating the transport component of the prehensile movement. For the grasp-preshape component, both peak aperture and peak-aperture velocity increased as target velocity increased, regardless of condition, indicating that target velocity was used to control the spatial aspects of aperture formation. However, the timing of peak aperture was affected by target velocity in the distance-controlled condition, but not in the viewing-time-controlled condition. These results provide evidence for the autonomous generation of the spatial and temporal aspects of grasp preshape. Thus, an independence between the transport and grasp-preshape phases was found, whereby the use of target velocity as a source of information for generating the transport component was limited; however, target velocity was an important source of information in the grasp-preshape phase.

Grip forces when passing an object to a partner

The goal of the present study was to investigate how grip forces are applied when transferring stable control of an object from one person to another. We asked how grip forces would be modified by the passer to (1) control for inertial forces as the object was transported toward the receiver and (2) control for the impending perturbation when the receiver made contact with the object. Twelve volunteers worked in pairs during this experiment. One partner, playing the role of passer, transported an object with embedded load cells forward or held the object at an interception location. The second partner, playing the role of receiver, waited at an interception location or reached toward the passed object. Kinematic results indicated that while passers performed a stereotypical movement, receivers were sensitive to the motion of the object as they reached to make contact. Grip force results indicated that passers’ grip forces and grip/load force ratios were variable on a trial-to-trial basis, suggesting that a refined internal model of the passing task was not achieved within the timeframe of the experiment. Furthermore, a decoupling of the temporal and magnitude characteristics of the grip and inertial forces was noted in conditions where passers transported the object toward the receiver. During object transfer, it was noted that passers used visual feedback-based anticipatory control to precisely time initial grip force release, while somatosensory control was used by both the passer and receiver to precisely coordinate transfer rate.

Manual asymmetries in bimanual prehension tasks: manipulation of object size and object distance.

Two experiments were designed to investigate the temporal and spatial couplings of the transport and grasp components for bimanual movements to both congruent and incongruent targets. We studied conditions where task requirements were largely different for the two hands. Ten participants performed Experiment 1 and were required to reach for, grasp, and lift two small (1 mm) cylinders, two large (70 mm) cylinders, or one small and one large cylinder with the right and left hands. In Experiment 2, 10 participants were required to reach for, grasp, and lift two objects that were positioned either near (50mm) the start mark, far (maximum comfortable reaching distance) from the start mark, or one near and one far from the start mark. Kinematic measures, relative timing differences between the hands and spatial plots were used to quantify both temporal and spatial couplings of the limbs. For temporal coupling, the results from both experiments indicated that the upper limbs were controlled independently with some execution-level interference occurring for the transport component only. In terms of spatial coupling our results indicated weak coupling of the grasp component regardless of task parameters (i.e., congruent or incongruent movements) and a dependence on task parameters in determining the level of spatial coupling for the transport component. These results can be collectively interpreted as evidence for a functional coupling of the upper limbs. That is, the movements of the hands may be coupled during tasks in which temporal and spatial synchronizations are beneficial for performance. However, if the coupling of the upper limbs is either unimportant or perhaps even detrimental to the coordination of the overall movement, then the upper limbs may perform the desired movements independently of one another.

Temporal coordination during bimanual reach-to-grasp movements: the role of vision.

The performance of bimanual movements involving separate objects presents an obvious challenge to the visuo-motor system: Visual feedback can only be obtained from one target at a time. To overcome this challenge overt shifts in visual attention may occur so that visual feedback from both movements may be used directly (Bingham, Hughes, & Mon-Williams, 2008; Riek, Tresilian, Mon-Williams, Coppard, & Carson, 2003). Alternatively, visual feedback from both movements may be obtained in the absence of eye movements, presumably by covert shifts in attention (Diedrichsen, Nambisan, Kennerley, & Ivry, 2004). Given that the quality of information falls with increasing distance from the fixated point, can we obtain the level of information required to accurately guide each hand for precision grasping of separate objects without moving our eyes to fixate each target separately? The purpose of the current study was to examine how the temporal coordination between the upper limbs is affected by the quality of visual information available during the performance of a bimanual task. A total of 11 participants performed congruent and incongruent movements towards near and/or far objects. Movements were performed in natural, fixate-centre, fixate-left, and fixate-right vision conditions. Analyses revealed that the transport phase of incongruent movements was similar across vision conditions for the temporal aspects of both the transport and grasp, whereas the spatial aspects of grasp formation were influenced by the quality of visual feedback. We suggest that bimanual coordination of the temporal aspects of reach-to-grasp movements are not influenced solely by overt shifts in visual attention but instead are influenced by a combination of factors in a task-constrained way.

Coordination and concurrency in bimanual rotation tasks when moving away from and toward the body

In the present series of experiments we investigated how object transport and rotate movements are performed when they are directed away from (Experiment 1) and toward (Experiment 2) the body under both unimanual and bimanual conditions. Our results indicated that unimanual conditions are faster and more efficiently produced than bimanual movements in far peripersonal space, suggesting that there is a cost to performing bimanual movements. However, in near peripersonal space, bimanual same movements were performed in a manner similar to unimanual movements, indicating that there is no significant cost associated with similar bimanual movements that are performed using the lower visual field and in near peripersonal space. Both experiments also indicate that the two hands are tightly synchronized when the two movements being performed require the same rotation. However, when performing bimanual movements where the rotation being performed by the two hands is different, this synchronization is weaker. Finally, the combined results from the two experiments indicated that movements made toward the body are not performed in a similar manner to movements that are made away from the body. Specifically, it is clear from the current studies that movements toward the body are performed faster and possibly that the hands are less synchronized for bimanual movements requiring different rotations by the two hands.

Performance of unimanual and bimanual multiphased prehensile movements.

By manipulating task action demands in 2 experiments, the author investigated whether the context-dependent effects seen in unimanual multiphase movements are also present in bimanual movements. Participants (N = 14) in Experiment 1 either placed or tossed objects into targets. The results indicated that the intention to perform a subsequent action with an object could influence the performance of an earlier movement in a sequence in both unimanual and bimanual tasks. Furthermore, assimilation effects were found when the subsequent tasks being performed by the 2 hands were incongruent. In Experiment 2, the author investigated in 12 participants whether planning in a multiphase movement includes some representation of the accuracy demands of the subsequent task. The accuracy demands of a subsequent task did not appear to influence initial movement planning. Instead, the present results support the notion that it is the action requirements of the subsequent movement that lead to context-dependent effects.

An experimental study on the role of graphical information about hand movement when interacting with objects in virtual reality environments

In this series of experiments, we investigated whether a crude representation of the hand that was extinguished at movement onset improved performance when compared to a no-feedback situation. Subjects performed simple reach to grasp movements in a virtual environment in two experiments. In Experiment 1, trials were blocked so that subjects were aware that a graphical representation of the hand would either be available throughout the movement (FA), be removed at movement onset (FAB), or not be available (NF). In Experiment 2, trials were randomized so that subjects were unaware of whether feedback would be available throughout the trial or removed at movement onset. Our results indicated that when subjects were aware of the availability of graphical feedback, the FAB condition improved performance compared to the NF condition. Furthermore, movement time was similar in the two feedback available conditions (FA, FAB). In contrast, for the randomized trial presentation, the positive influence of the FAB condition was diminished. These results suggest that visual feedback available prior to movement onset can be used to calibrate the proprioceptive system and improve performance over a no feedback situation. These results can be applied by designers of virtual environments to solve problems related to occlusion of important environmental information by the hand as users reach to grasp and manipulate objects.

The effects of visual information about self-movement on grasp forces when receiving objects in an augmented environment

This work explored how the presence of visual information about self-movement affected grasp forces when receiving an object from a partner. Twelve subjects either reached to grasp or grasped without reaching objects that were passed by a partner or rested on a table surface. Visual feedback about self-movement was available for half the trials and was removed for the other half. Results indicated that a graphic representation of self-movement significantly decreased transfer time when objects were passed between subjects. Results also indicated decreased time to peak grip force and peak grip force rate by the receiver with this visual feedback. These results suggest that grip force production on objects acquired from another person benefit from a crude graphical representation of the finger pads. Furthermore, these results suggest that sources of sensory feedback cannot be studied in isolation. Instead we must consider how feedback modalities are integrated for successful interaction. Implications for the design of virtual environments and integrated feedback devices are discussed.

Age differences in the control of a precision reach to grasp task within a desktop virtual environment

The purpose of this experiment is to investigate the fine motor performance of young and older adults on a reach to grasp task in a desktop virtual environment with increasing precision requirements. Aging brings about potential loss of an individuals function due to disease, injury, or the degenerative nature of aging itself. Three-dimensional virtual environments have been identified as systems with good potential to ameliorate such problems in older individuals, and precise fine motor skills represent an important class of functional skills. Two groups of participants (Young, n=10, mean age 21.3 years, range 20-24, senior, n=10, mean age 70.7 years, range 60-85) performed a reach to grasp in a desktop virtual environment with simple, low contrast graphics. Results indicate that visual feedback of the hand for sensory guidance of movement did not improve motor performance for either group, and that as precision requirements of the task increased, age group differences in movement time and peak grasp aperture also increased. These findings extend the literature on age group differences in human motor control across the lifespan and differ from previous studies which showed presence of visual feedback of the hand improved motor performance in young adults. Differences in luminance contrast levels in past studies and the current one suggest that control over this feature of the visual scene is an important design consideration for all end-users and warrants additional investigation. Additional recommendations for age-specific design of three dimensional user interfaces include usage of tangibles that are sufficient in size to limit detrimental effects for older adults.