Supervised and unsupervised learning of multidimensional acoustic categories.

Learning to recognize the contrasts of a language-specific phonemic repertoire can be viewed as forming categories in a multidimensional psychophysical space. Research on the learning of distributionally defined visual categories has shown that categories defined over 1 dimension are easy to learn and that learning multidimensional categories is more difficult but tractable under specific task conditions. In 2 experiments, adult participants learned either a unidimensional or a multidimensional category distinction with or without supervision (feedback) during learning. The unidimensional distinctions were readily learned and supervision proved beneficial, especially in maintaining category learning beyond the learning phase. Learning the multidimensional category distinction proved to be much more difficult and supervision was not nearly as beneficial as with unidimensionally defined categories. Maintaining a learned multidimensional category distinction was only possible when the distributional information that identified the categories remained present throughout the testing phase. We conclude that listeners are sensitive to both trial-by-trial feedback and the distributional information in the stimuli. Even given limited exposure, listeners learned to use 2 relevant dimensions, albeit with considerable difficulty. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The control of sequential goal-directed movement: learning to use feedback or central planning?

Fifteen participants practiced a two-target sequential aiming movement with either full vision of the movement environment, vision during flight, or vision while in contact with the first target. After 100 acquisition trials, participants performed a retention test in their own condition and then were transferred to each of the other two vision conditions. Both performance and kinematic data indicated that rather than becoming less dependent on visual information with practice, subjects learned to adjust their movement trajectories to use the visual information available in their particular vision condition. Although transfer to a degraded vision condition disrupted performance, when vision was augmented participants quickly adjusted their aiming trajectories to use the added information. The findings suggest that at least part of learning involves the development of rapid and efficient procedures for processing afferent information, including visual response-produced feedback.     

Comparing exemplar and prototype models of categorization.

Five experiments contrasted the prototype and exemplar theories of categorization. The stimuli for the experiments were simple perceptual figures that varied along two multi-valued dimensions. During the learning phase of each experiment, participants assigned each of a series of stimuli to one of two categories and received feedback concerning their assignments. During the generalization phase, participants made category judgments with respect to novel stimuli, including some crucial test stimuli that were (a) more similar to the prototype of one of the two categories, and at the same time, (b) more similar, on average, to the exemplars of the other category. Participants assigned the test stimuli to the latter category. This was taken as evidence favoring the exemplar over the prototype theory. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Reduced feedback frequency enhances generalized motor program learning but not parameterization learning.

Examined the effects of a reduced feedback frequency on the learning of generalized motor programs and movement parameterization. Ss practiced 3 movement patterns with the same relative timing and the same relative amplitude, but with varied movement time (Exp 1) or varied movement amplitude (Exp 2). Knowledge of results (KR) was given either on 100% or 63% of the trials, with learning being assessed by retention and transfer tests. In both experiments, reduced KR frequency enhanced generalized motor program (GMP) learning but generally degraded parameter learning. These data provide converging evidence for the dissociation of the program and parameterization processes postulated in GMP theory. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Proprioceptive Perception of Phase Variability.

Previous work has established that judgments of relative phase variability of 2 visually presented oscillators covary with mean relative phase. Ninety degrees is judged to be more variable than 0° or 180°, independently of the actual level of phase variability. Judged levels of variability also increase at 180°. This pattern of judgments matches the pattern of movement coordination results. Here, participants judged the phase variability of their own finger movements, which they generated by actively tracking a manipulandum moving at 0°, 90°, or 180°, and with 1 of 4 levels of Phase Variability. Judgments covaried as an inverted U-shaped function of mean relative phase. With an increase in frequency, 180° was judged more variable whereas 0° was not. Higher frequency also reduced discrimination of the levels of Phase Variability. This matching of the proprioceptive and visual results, and of both to movement results, supports the hypothesized role of online perception in the coupling of limb movements. Differences in the 2 cases are discussed as due primarily to the different sensitivities of the systems to the information. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Perceptual learning immediately yields new stable motor coordination.

Coordinated rhythmic movement is specifically structured in humans. Movement at 0° mean relative phase is maximally stable, 180° is less stable, and other coordinations can, but must, be learned. Variations in perceptual ability play a key role in determining the observed stabilities so we investigated whether stable movements can be acquired by improving perceptual ability. We assessed movement stability in Baseline, Post Training, and Retention sessions by having participants use a joystick to coordinate the movement of two dots on a screen at three relative phases. Perceptual ability was also assessed using a two-alternative forced choice task in which participants identified a target phase of 90° in a pair of displays. Participants then trained with progressively harder perceptual discriminations around 90° with feedback. Improved perceptual discrimination of 90° led to improved performance in the movement task at 90° with no training in the movement task. The improvement persisted until Retention without further exposure to either task. A control group's movement stability did not improve. Movement stability is a function of perceptual ability, and information is an integral part of the organization of this dynamical system. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Effects of aging on planning and implementing arm movements.

In Exps 1 and 2, aiming movements were performed with and without visual feedback in young and elderly adults. The initial (acceleration and deceleration phases) and secondary movement components were analyzed. Although deceleration phase accuracy decreased without visual feedback in both age groups, accuracy diminished as movement amplitude increased only in the elderly. This suggested that the elderly were more dependent on visual feedback to modify motor programs for longer duration movements. Velocity also increased less with increasing amplitude and target size in the elderly, which was related to impaired preprogramming (acceleration phase) and implementation (deceleration phase) of higher forces. This conclusion was confirmed directly in Exp 2 because only the deceleration phase was affected by the removal of a visual feedback of arm position when availability of visual information could not be predicted before movement. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Transfer effects of cue-related movement reversal in discrimination reversal learning with rhesus monkeys.

Investigated the relative influence of directional movement reversal on the learning of discriminations and discrimination reversals using 36 macaque monkeys in 2 experiments. The learning tasks involved comparisons among movement feedback, brightness, and spatial location as discriminanda. Results indicate that differential directional movements, when related to or serving as discriminative cues, produce a persistent elevated level of postreversal errors within a discrimination reversal task. Further, the brightness dimension was relatively easy for reversal and generally superior to spatial location and movement feedback. It is suggested that the transfer difficulty produced by cue-related differential movement might be attributed to proactive interference from the reorganization of response-response sequences within the response. (16 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The effect of type and timing of feedback on learning from multiple-choice tests.

Two experiments investigated how the type and timing of feedback influence learning from a multiple-choice test. First, participants read 12 prose passages, which covered various general knowledge topics (e.g., The Sun) and ranged between 280 and 300 words in length. Next, they took an initial six-alternative, multiple-choice test on information contained in the passages. Feedback was given immediately for some of the multiple-choice items or after delay for other items. Participants were either shown the correct answer as feedback (standard feedback) or were allowed to keep answering until the correct answer was discovered (answer-until-correct feedback). Learning from the test was assessed on a delayed cued-recall test. The results indicated that delayed feedback led to superior final test performance relative to immediate feedback. However, type of feedback did not matter: discovering the correct answer through answer-until-correct feedback produced equivalent performance relative to standard feedback. This research suggests that delaying the presentation of feedback after a test is beneficial to learning because of the spaced presentation of information. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Acquiring bimanual skills: Contrasting forms of information feedback for interlimb decoupling.

Addressed the learner's capability to perform different upper-limb actions simultaneously with the help of various sources of information feedback. An elbow flexion movement was made in the left limb together with a flexion-extension-flexion movement in the right limb. Interlimb interactions were assessed at the structural as well as the metrical level of movement specification during acquisition and retention. Despite a strong initial tendency for the limbs to be synchronized, findings revealed that Ss became gradually more successful in interlimb decoupling as a result of practice with augmented feedback. However, detailed knowledge of movement kinematics was no more effective than global outcome information for interlimb decoupling, indicating that knowledge of results may have more potential for acquiring multiple degree-of-freedom tasks than previously believed. Finally, the data support the general notion that learning new coordination tasks involves the suppression of preexisting preferred coordination tendencies, which is often a prerequisite for building new coordination modes. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Information-Integration Category Learning in Patients With Striatal Dysfunction.

Information-integration category learning was examined in patients with Parkinson's disease (PD) and in healthy control participants in 2 different conditions. In the linear condition, optimal categorization required a nonverbalizable linear integration of information from the 2 stimulus dimensions, whereas in the nonlinear condition, a nonlinear integration of information was required. Each participant completed 600 trials in each condition and was given corrective feedback following each trial. Results indicated that PD patients were not impaired in the linear condition across all trials, whereas the same patients were impaired in the nonlinear condition, but only later in training. The authors conducted model-based analyses to identify participants who used an information-integration approach, and a comparison of the accuracy rates of those individuals further revealed a specific deficit in information-integration category learning in patients with PD. These findings suggest that the striatum may be particularly involved in information-integration category learning when the rule is highly complex. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Impact of Perceived Consensus on Stereotypes About Obese People: A New Approach for Reducing Bias.

In 3 experiments, the authors tested the effect of perceived social consensus on attitudes toward obese people. Participants completed self-report measures of attitudes toward obese people prior to and after manipulated consensus feedback depicting attitudes of others. In Study 1 (N=60), participants decreased negative and increased positive stereotypes after learning that others held more favorable attitudes toward obese people. In Study 2 (N=55), participants improved attitudes when they learned about favorable attitudes of obese people from an in-group versus an out-group source. In Study 3 (N=200), a consensus approach was compared with other stigma reduction methods. Social consensus feedback influenced participants' attitudes and beliefs about causes of obesity. Providing information about the uncontrollable causes of obesity and supposed scientific prevalence of traits also improved attitudes. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The role of visual and nonvisual feedback in a vehicle steering task.

This article investigates vehicle steering control, focusing on the task of lane changing and the role of different sources of sensory feedback. Participants carried out 2 experiments in a fully instrumented, motion-based simulator. Despite the high level of realism afforded by the simulator, participants were unable to complete a lane change in the absence of visual feedback. When asked to produce the steering movements required to change lanes and turn a corner, participants produced remarkably similar behavior in each case, revealing a misconception of how a lane-change maneuver is normally executed. Finally, participants were asked to change lanes in a fixed-based simulator, in the presence of intermittent visual information. Normal steering behavior could be restored using brief but suitably timed exposure to visual information. The data suggest that vehicle steering control can be characterized as a series of unidirectional, open-loop steering movements, each punctuated by a brief visual update. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The source of execution-related dual-task interference: Motor bottleneck or response monitoring?

The present study assessed the underlying mechanism of execution-related dual-task interference in the psychological refractory period (PRP) paradigm. The motor bottleneck hypothesis attributes this interference to a processing limitation at the motor level. By contrast, the response monitoring hypothesis attributes it to a bottleneck process that not only selects the appropriate response but also monitors its execution. In two experiments, participants performed ballistic movements of different distances in Task 1 and a choice reaction time task in Task 2. In each experiment, a propagation effect of movement distance on reaction time in Task 2 indicated substantial execution-related interference. To determine the locus of this effect, we manipulated stimulus-response compatibility in Task 2. In line with the motor bottleneck hypothesis, the compatibility effect was partially absorbed during movement execution of Task 1. The results support a motor bottleneck mechanism rather than response monitoring as the source of execution-related dual-task interference. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The effect of the Müller-Lyer illusion on the planning and control of manual aiming movements.

Two experiments used Müller-Lyer stimuli to test the predictions of the planning-control model (S. Glover, 2002) for aiming movements. In Experiment 1, participants aimed to stimuli that either remained the same or changed upon movement initiation. Experiment 2 was identical except that the duration of visual feedback for online control was manipulated. The authors found that the figures visible during movement planning and online control had additive effects on endpoint bias, even when participants had ample time to use visual feedback to modify their movements (Experiment 2). These findings are problematic not only for the planning-control model but also for A. D. Milner and M. A. Goodale's (1995) two visual system explanation of illusory bias. Although our results are consistent with the idea that a single representation is used for perception, movement planning, and online control (e.g., V. H. Franz, 2001), other work from our laboratory and elsewhere suggests that the manner in which space is coded depends on constraints associated with the specific task, such as the visual cues available to the performer. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

Integrating Novel Dimensions to Eliminate Category Exceptions: When More Is Less.

Category learning can be characterized as a process of discovering the dimensions that represent stimuli efficiently and effectively. Categories that are overlapping when represented in 1 dimensionality may be separate in a higher dimensional cue set. The authors report 2 experiments in which participants were shown an additional cue after learning to use 2 imperfect cues. The results revealed that participants can integrate new information into their categorization cue set. The authors discovered wide individual differences, however, with many participants favoring simpler, but less accurate, cue sets. Some participants demonstrated the ability to discard information previously used when new, more accurate information was introduced. The categorization model RASHNL (J. K. Kruschke & M. K. Johansen, 1999) gave qualitatively accurate fits of the data. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Why is the alphabetically middle letter in a multiletter array so hard to determine? Memory processes in linear-order information processing

In 3 experiments, participants identified target letters in multiletter strings of different interletter steps. Results show that the alphabetically extreme targets in the arrays (A or C in BCA) were consistently identified faster than were the interior letters, with the alphabetically middle target (B in BCA) identified most slowly. Distance effects were not consistent across different targets. The findings are not consistent with the ends-inward serial search, comparison, and serial search plus comparison models. The data suggest that an interior target was identified by first identifying the terminal letter and then mentally constructing the linear order inward by probabilistically using either a comparison or an interitem serial search process to achieve maximal efficacy. This suggestion is consistent with the ends-inward construction theory of linear-order learning.     

The nature of phoneme representation in spoken word recognition.

Four experiments used the psychological refractory period logic to examine whether integration of multiple sources of phonemic information has a decisional locus. All experiments made use of a dual-task paradigm in which participants made forced-choice color categorization (Task 1) and phoneme categorization (Task 2) decisions at varying stimulus onset asynchronies. In Experiment 1, Task 2 difficulty was manipulated using words containing matching or mismatching coarticulatory cues to the final consonant. The results showed that difficulty and onset asynchrony combined in an underadditive way, suggesting that the phonemic mismatch was resolved prior to a central decisional bottleneck. Similar results were found in Experiment 2 using nonwords. In Experiment 3, the manipulation of task difficulty involved lexical status, which once again revealed an underadditive pattern of response times. Finally, Experiment 4 compared this prebottleneck variable with a decisional variable: response key bias. The latter showed an additive pattern of responses. The experiments show that resolution of phonemic ambiguity can take advantage of cognitive slack time at short asynchronies, indicating that phonemic integration takes place at a relatively early stage of spoken word recognition. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Can performance-feedback accuracy be improved? Effects of rater priming and rating-scale format on rating accuracy.

Performance appraisal information is often used for employee feedback and development. Research has found that assessments that are global (i.e., based on broad aspects of performance) and comparative (i.e., explicit interratee comparisons) may be most accurate in terms of Cronbach's (1955) differential accuracy, a type of accuracy that is directly relevant to the provision of feedback. Unfortunately, a global-comparative assessment may not give recipients the most useful diagnostic feedback. In this experiment, an innovative rater-priming manipulation was developed and tested on a sample of 109 participants. The priming manipulation had the effect of improving differential accuracy and providing diagnostic feedback. A 2nd independent variable involving 2 different Behavioral Observation Scale formats also was investigated. Explanations of findings, limitations of this experiment, directions for future research, and implications for performance appraisal practice are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)