The ecological approach to perceiving as a foundation for understanding the development of knowing in infancy

The ecological thesis of direct realism is used as a framework for examining the development of knowing in human infancy. When information for perceiving is defined ecologically (i.e., relative to the situational context and to the physical dimensions, capabilities, and needs of the perceiver), knowing need not be construed as the act of using representations to give meaning to acts or percepts. Knowing, alternatively, is the act of noticing affordances, situation and perceiverspecific meanings of objects, according to their value to the perceiver for achieving specific goals. Changes with development in infant sensorimotor functioning may, in this view, be explained by a process of increasing economy in noticing potentially available affordances, rather than a process of constructing a representational system for making present something not present. Studies of three infant skills widely attributed to the onset of representation are examined with regard to this ecological thesis. Results indicate that the noticing of affordances is critically involved in each of these skills.     

Transfer of training: Effects of successive pretraining of components in a dimension-abstracted oddity task

An experiment was conducted investigating the effects of training two components of a dimension-abstracted oddity problem, oddity responding and attention. All Ss were given a series of three problems. Type of oddity training was manipulated on Problem 1, and Ss learned (1) an oddity response in a problem whose format was the same as the final transfer problem, (2) an oddity response in a problem whose format was different from the final problem, or (3) no oddity response. Attention training was manipulated on Problem 2. The results indicated that Ss given same format oddity training and appropriate attention training were able to integrate that information on the final problem, that the probability of transferring the oddity response learned on Problem 1 to Problem 3 depended upon the similarity of the task formats, and that the two trained components tended to combine in an interactive fashion.     

Rule learning in children: Multiple-set training and intermediate-size transposition

Two experiments were conducted in which the subjects were trained to choose the middle-size stimulus in either a single stimulus set or in each of two nonoverlapping stimulus sets. Following criterion, each subject was presented with a series of 12 test trials involving a new stimulus set. In Experiment 1, kindergarten children tended to choose the middle-size test stimulus if the test set consisted of stimuli intermediate in size to the stimuli in the smaller and larger training sets. When the test stimuli were all larger than any of the training stimuli, the probability of choosing the middle-size stimulus increased as the distance between the two training sets increased. Three groups of children participated in Experiment 2: a kindergarten group, a first-grade group, and a first-grade group that experienced middle-size training a year earlier. The distance between training sets and two training procedures affected the three groups in a similar manner during acquisition, but several interactions were obtained during the test. The findings were discussed in terms of the Lane and Rabinowitz (D. M. Lane & F. M. Rabinowitz, Child Development, 1977, 48, 412–426) rule based theory of intermediate-size transposition and suggestions were made as to how children acquire the concept of middle.     

The Boston Naming Test: Norm data and cue utilization in a sample of normal 6- and 7-year-old children

The object-naming abilities of 83 normal children were examined using the Boston Naming Test developed by E. Kaplan, H. Goodglass, and S. Weintraub (1976, The Boston Naming Test (experimental edition), unpublished test, Veterans Hospital, Boston, Massachusetts). There is currently a lack of available norms for this instrument, especially for populations of younger children. Cue utilization in this sample is explored and attention is focused upon possible gender differences in naming performance and cue strategies. Seven summary scores derived from the test are presented in this report. Results indicate that the Boston Naming Test can discriminate within a normal population sample of young children and provides a relatively normal distribution for six of seven derived scores. Boys were found to have a significantly greater number of items correctly named but no gender differences in cue utilization were evident. A major proportion of the children in this sample of 6- and 7-year-olds used phonological cues almost exclusively, rather than semantic cues also provided. Cue use was not related to age or spontaneously given responses, and none of the children were found to be highly proficient at using both types of cues.     

A comparison of self-directed and therapist-directed practice for fear reduction

Various amounts of self-directed and therapist-directed practice were evaluated in reducing fear of snakes. The conditions included an exclusively self-directed condition, a predominantly self-directed condition, a predominantly therapist-directed condition, and an exclusively therapist-directed condition. All treatments resulted in reduced fear on behavioral and self-report measures. The predominantly therapist-directed subjects showed more improvement than subjects in the exclusively self-directed and predominantly self-directed conditions, but not more improvement than subjects in the exclusively therapist-directed condition.     

Effects of positive and negative adult-child interactions on children's social preference

The effects of positive and negative interactions on children's performance and preferences for the adults associated with each type of interaction were studied. Five children served as subjects during daily sessions in which each of three adults followed prescribed patterns of social interaction. One adult (positive) dispensed positive comments contingent upon either color-sorting or completion of arithmetic problems; a second adult (negative) mildly reprimanded the child for off-task behavior; and the third adult (extinction or nonreactive) said nothing to the child when he was present. Following each session the three adults reentered and the child chose one of them for an additional period of interaction. Three response measures (frequency, latency, and percent of time on-task) and children's adult preferences were obtained daily. The negative adult affected the most task behavior and had strongest stimulus control; the positive adult, though exerting little control over children's behavior, was the most preferred. Children's adult preferences were not specific to the task setting, but were observed across a variety of contexts.     

Patterns of spontaneous recovery in aphasic groups: A study of adult stroke patients

The recovery of 31 aphasic stroke patients in the first three months postonset was investigated. Improvements on eight comprehension, repetition, and expression tasks were documented for four groups. These groups, retrospectively defined on the basis of initial scores on a fluency and a comprehension task, were: Low Fluency/Low Comprehension (LFLC), Low Fluency/High Comprehension (LFHC), High Fluency/Low Comprehension (HFLC), and High Fluency/High Comprehension (HFHC). Examination of each group's recovery revealed (1) equal improvement on nearly all language tasks for the two High Comprehension groups and (2) a more selective improvement largely in comprehension and imitation tasks for the two Low Comprehension groups. Implications of this selective improvement in initially low comprehending aphasics are discussed.     

Contribution of the speech musculature to apparent human EEG asymmetries prior to vocalization

Vocalizations are preceded by slow EEG waves over inferior frontal scalp regions, but some controversy exists as to whether these slow potentials are left-lateralized as a function of speech. In this study both speech and non-speech vocalizations were studied, under conditions of repetitive and variable production. Averaged responses over the speech musculature, including the temporalis, masseter, orbicularis oris, and mylohyoid muscles, as well as the eyes and tongue, were calculated simultaneously with scalp EEG averages. The responses of the temporalis and masseter muscles, as well as that of the tongue, were correlated with EEG sites at specific latencies prior to speech, suggesting that no part of the inferior frontal EEG record was free from myogenic confounds. In addition, all significant EEG differences between speech and non-speech conditions were subsequently eliminated by covariance analysis, using selected muscle channels as covariates. Previous reports of asymmetrical “speech potentials” in the EEG probably reflected a combination of several myogenic confounds.     

The effect of continuous, intermittent, and "placebo" reinforcement on the effectiveness of the conditioning treatment for enuresis nocturna

It has been suggested that intermittent reinforcement may be of value in reducing the rate of relapse in bedwetting conditioning therapy. An experiment was conducted with 30 enuretic boys to examine acquisition and extinction parameters of continuous (100%), intermittent (70% variable ratio), and “placebo” (0%) reinforcement schedules. Reinforced trials were administered as in the typical Mowrer conditioning procedure. Non-reinforced trials were achieved by means of a time delay with the subsequent alarm being activated in the parent's room rather than in the child's room. These procedures required the development of a new conditioning device which could be programmed to automatically administer the desired schedule of reinforcement. The results of the field investigation revealed that continuous reinforcement (CR) and intermittent reinforcement (IR) groups attained acquisition in approximately the same number of trials and with essentially the same success rate. Relapse rate was significantly greater in the CR group than in the IR group. The placebo group showed no improvement over the 6 weeks of treatment. The results of this study may be interpreted as supporting the tenet that relapse can be viewed and treated as an extinction of the acquired response.     

Affordances and infant learning: A reply to Horowitz

We reply to three major points made by F. Horowitz (1983, Developmental Review, 3, 405–409) in her commentary on the ecological approach to infant knowing presented by E. Goldfield (1983, Developmental Review, 3, 371–404). We first clarify the relation between perceiving and acting from an ecological perspective, and distinguish between affordances as environmental properties scaled to the perceiver/performer and representations as mental structures. We then present a critique of the process of association offered by Horowitz as an explanation of infant learning. Association fails to specify the constraints on what is learned, while the ecological process of noticing affordances, presented by Goldfield, considers such constraints. We conclude by presenting operational criteria for measuring affordances and evidence that perception is scaled to the perceiver/performer.     

A deletion ahead of its time

An alleged defect in transformational treatment of syntax acquisition is the absence from child speech of certain predicted errors with ‘wh-’ constructions. In this paper, a theory of acquisition dynamics and intensive longitudinal data are brought to bear on this issue. The key observations involve an early precursor, at 24—28 months, of the relative clause. The analysis sheds light on two fundamental issues in transformational acquisition theory: The permissibility of simultaneous rule changes and whether a transformation can be acquired before the associated deep structure. The issues and analysis can be translated into non-transformational terms.