Michael K. Gardner

Twenty‐year outcome for individuals with autism and average or near‐average cognitive abilities

Previous studies found substantial variability in adult outcome for people with autism whose cognitive functioning was within the near‐average and average ranges. This study examined adult outcome for 41 such individuals (38 men and 3 women) originally identified through an epidemiological survey of autism in Utah. Mean age at the time of their previous cognitive assessment was 7.2 years (SD=4.1, range=3.1–25.9 years) and at follow‐up was 32.5 years (SD=5.7 years, range=22.3–46.4 years). Outcome measures included standardized assessments of diagnostic status, cognitive ability, and adaptive behavior. Additional information collected concerned demographic variables, indicators of independence, social relationships, medical and psychiatric conditions, and social service use. Outcomes for this sample were better than outcomes described in previous work on individuals with similar cognitive functioning. For example, half of the participants were rated as “Very Good” or “Good” on a global outcome measure. As in previous studies, there was considerable variability in measured cognitive ability over time. Over half of the sample had large gains or losses of cognitive ability of greater than 1 standard deviation. Cognitive gain was associated with better outcome, as was better adaptive functioning. While all participants had baseline IQs in the nonimpaired range, there was limited evidence to support the use of other early childhood variables to predict adult outcome.

Negative transfer errors in sequential cognitive skills: strong-but-wrong sequence application.

Three experiments investigated the role of processing sequence knowledge in negative transfer within multistep cognitive skills. In Experiments 1 and 2, more training resulted in higher error rates when new processing sequences that resembled familiar ones were introduced in transfer. Transfer error responses were executed with the same speed as correct responses to familiar sequence trials, and the errors appeared to be undetected by the performers. Experiment 3 tested whether the effects of sequence learning were attributable to explicit or implicit knowledge of processing sequences. Evidence favored the implicit learning interpretation. Findings are discussed in relationship to earlier demonstrations of the einstellung effect and to current taxonomic theories of human error.

The word frequency effect in lexical decision: finding a frequency-based component

Subjects making lexical decisions are reliably faster in responding to high-frequency words than to low-frequency words. This is known as the word frequency effect. We wished to demonstrate that some portion of this effect was due to frequency differences between words rather than to other dimensions correlated with word frequency. Three groups of subjects (10 engineers, 10 nurses, and I0 law students) made lexical decisions about 720 items, half words and half nonwords, from six different categories (engineering, medical, low-frequency nontechnical, medium-frequency nontechnical, and two groups of high-frequency nontechnical). Results of two analyses of variance revealed a crossover interaction such that engineers were faster in responding to engineering words than to medical words, whereas nurses were faster in responding to medical words than to engineering words. The engineering and medical words were equally long and equally infrequent by standard word counts. We take this as support for a frequency-based component in the word frequency effect. The practical implications of this research for estimating the readability of technical text are discussed.

Childhood traumatic brain injury: neuropsychological status at the time of hospital discharge

The present study examined the status at the time of hospital discharge of 22 children who had sustained moderate to severe closed head injuries. Despite an average length of stay of 46 days, on average the children performed significantly below normal on cognitive, memory, and motor tests. The children who did the worst were those who sustained their injuries riding in a vehicle, as opposed to being hit by a vehicle while walking or riding a bicycle. No reliable associations were found between neuroimaging data and cognitive measures; however, children whose scans showed more diffuse severe injury (e.g. white matter hemorrhages) performed more poorly than those who did not. Of further importance were detailed morphometric analyses of tissue quantity and ventricular volume that showed significantly increased ventricle to brain ratios (VBR) over a 3 to 4 week period following injury.

The role of attention processes in near transfer of cognitive skills

We tested hypotheses about the respective roles of working memory, perceptual speed, knowledge, and attention disengagement processes in skill transfer errors that resemble einstellung effects or strong-but-wrong slips. Correlational evidence was consistent with the hypothesis that these errors are primarily due to the inability of learners to disengage attention from expected or familiar operations. The data also supported the notion that error proneness during near transfer generalizes across different skills. Contrary to expectations, working memory capacity had little or no relationship to either transfer errors or attention disengagement ability. Results are discussed with respect to skill training and individual differences in skill performance ability.

Individual differences in undetected errors in skilled cognitive performance

Abstract Skilled performers of cognitive tasks make fewer errors than novices, but they continue to make some errors. A number of these errors, which occur in near-transfer situations, go undetected by the performer. Such errors can have disastrous consequences. We investigated a number of individual differences variables that were correlated with undetected error making in a laboratory task (number reduction) requiring subjects to apply reduction rules to strings of digits to arrive at a single digit answer. In two experiments, we determined: (a.) self-report measures of the propensity to make cognitive slips and lapses were statistically unrelated to performance on our experimental task; and (b) performance-based cognitive resource measures (i.e., working memory capacity) were related to performance. Self-report measures appear to be influenced by social desirability or some correlated construct.

Changes in analogical reasoning in adulthood

This study sought to investigate adult intellectual development through an analysis of a particular type of cognitive ability, verbal analogical reasoning. The performance of 60 individuals between the ages of 20 and 79 was compared on 100 verbal analogies. The subjects consisted of six groups of ten individuals each (five males and five females), matched as a group for education and gender. Solution times and error rates served as the dependent measures. Results showed that there was a significant trend for the older subjects (60- and 70-year-olds) to be slower than the young subjects (20-, 30-, 40-, and 50-year-olds), but not necessarily more error prone. These data suggest that verbal analogical reasoning changes with age. Supplemental data demonstrated a change in other abilities as well (i.e., decline in perceptual-motor speed and spatial skill).

Effectiveness of Stimulant Drug Treatment for Attention Problems: A Look at Head-Injured Children

Using a within-subject, repeated measure, double-blind design, methylphenidate was shown to be effective in treating the attention deficits of children with head injuries. Eight head-injured children, ages 7-15, meeting DSM III-R criteria for ADHD, were placed on a 2-week trial of methylphenidate (Ritalin), and similar 2-week trial of placebo. During the methylphenidate trial, all eight children were rated by parents as improved in their attentional capacities, and all but one, a child diagnosed as ADHD prior to injury and described as conduct-disordered, showed a pattern of general improvement across most attentional measures. Measures which showed less consistent results were the attentional tasks comprised of more complex stimuli requiring greater memory capacity. Implications for school psychologists working with head-injured children are discussed.

Relationship between Perceived and Actual Frequency Represented by Common Rating Scale Labels.

Two experiments investigated the relationship between subjective interpretation of frequency terms and corresponding objective values. Evidence supported the existence of a nonlinear relationship that is well described by a logarithmic function. The general form of this relationship was consistent across different methods of eliciting subjective frequency estimates and different frequency response scales. These findings, in combination with prior research, suggest that individuals can be highly accurate in reporting the frequency of prior experiences but that subjective reports using common frequency response scales and scoring methods should not be taken to reflect equal intervals of actual frequency. This bears similarity to psychophysical relationships observed between sensation and stimulus intensity, and it has implications for the interpretation and analysis of self-report measures that use Likert-type frequency response scales.

A Procedural Approach to Remembering Personal Identification Numbers among Older Adults

This study investigated whether a motor skill learning intervention could provide better memory for personal identification numbers (PINs) as compared to a control group. Younger (ages 18 to 40) and older (ages 61 to 92) participants were randomly assigned to conditions. All participants received three days of training consisting of 12 blocks of 12 trials each. Participants were tested immediately after training, after four days, and after seven days. Dependent measures were errors, latencies, and number of correct responses per minute. Younger participants were less error prone, faster, and produced more correct responses than older participants. Training condition (motor skill-based versus control training) had no significant effect on any of the dependent variables. Testing time had a significant effect on latency, and the effect of testing time on latency interacted with age group. In a second study, six older individuals diagnosed as having mild cognitive impairment (MCI) were trained using the motor skill learning intervention. Their performance was compared with that of the younger and older motor skill groups from the first experiment. The results showed that the older MCI group was significantly slower, more error prone, and produced fewer correct responses per minute than the older, normal group. Thus the presence of diagnosed MCI significantly impairs memory for PINs beyond the impairment expected from normal aging.