Sandra Hale

Processing Speed, Working Memory, and Fluid Intelligence: Evidence for a Developmental Cascade

Processing speed, working memory capacity, and fluid intelligence were assessed in a large sample (N = 214) of children, adolescents, and young adults (ages 7 to 19 years) Results of path analyses revealed that almost half of the age-related increase in fluid intelligence was mediated by developmental changes in processing speed and working memory and nearly three fourths of the improvement in working memory was mediated by developmental changes in processing speed Moreover, even when age-related differences in speed, working memory and fluid intelligence were statistically controlled, individual differences in speed had a direct effect on working memory capacity which, in turn was a direct determinant of individual differences in fluid intelligence

The Information-Loss Model: A Mathematical Theory of Age-Related Cognitive Slowing

A model of cognitive slowing is proposed with the following assumption: Information is lost during processing, processing occurs in discrete steps with step duration inversely related to the amount of information currently available, and the effect of aging is to increase the proportion of information lost per step. This model correctly predicts a positively accelerated relation between latencies of older and younger adults and provides a unified account of the effects of task complexity, practice, speed-accuracy tradeoffs, and fluctuations in individual performance. Strong support for the thesis that cognitive slowing is global, and not localized in specific age-sensitive components, is provided by the fact that the model accurately predicts the latencies of older adults on the basis of those of younger adults, without regard to the nature of the task, across a latency range of nearly 2 orders of magnitude.

Relationships among processing speed, working memory, and fluid intelligence in children

The present review focuses on three issues, (a) the time course of developmental increases in cognitive abilities; (b) the impact of age on individual differences in these abilities, and (c) the mechanisms by which developmental increases in different aspects of cognition affect each other. We conclude from our review of the literature that the development of processing speed, working memory, and fluid intelligence, all follow a similar time course, suggesting that all three abilities develop in concert. Furthermore, the strength of the correlation between speed and intelligence does not appear to change with age, and most of the effect of the age-related increase in speed on intelligence appears to be mediated through the effect of speed on working memory. Finally, most of the effect of the age-related improvement in working memory on intelligence is itself attributable to the effect of the increase in speed on working memory, providing evidence of a cognitive developmental cascade.

The rise and fall in information-processing rates over the life span

We surveyed studies that measured information-processing durations in groups of experimental subjects (children or elderly adults) and a group of college-aged control subjects. Some studies varied the type of processing while keeping the age of a subject group fixed. Process-durations in experimental subjects could be described by a multiplicative function of the control durations, regardless of the type of processing. Other studies varied the age of the subject groups while keeping the type of processing fixed. Process-durations declined during childhood, in a manner that could be described by a negative exponential function of age. Process-durations increased throughout middle- and old-age, in a manner that could be described by a positive exponential function of age. The sum of the two exponentials defined a U-shaped function that described process-durations over the life span. The most important studies varied both the type of processing and the age of the subject groups. An array of measurements of this kind could be described by a two-dimensional function that combined the multiplicative effect of process-duration and the exponential effects of age. The multiplicative effect of process-duration suggested that the execution of a processing sequence was conditioned by a single developmental parameter in both the experimental subject and the control subject. The exponential components determined the magnitude of the developmental parameter as the age of the subject changed. Given the global character of these effects, it seemed to us that the developmental mechanism may operate at a more elementary level than the information-processing stages conceived by cognitive theories. In a developmental framework, information processing may be reducible to a large number of small steps of a homogeneous duration or reliability, such as might be realized on a neural network. The exponential rate constants may be related to constant-probability hazards that act on one or another population of neural elements to create minute defects or incremental improvements. Their cumulative effects alter the functioning of the network over its lifetime, in a way that parallels the observed changes in process-durations.

Converging Evidence That Visuospatial Cognition Is More Age-Sensitive Than Verbal Cognition

In 3 separate experiments, the same samples of young and older adults were tested on verbal and visuospatial processing speed tasks, verbal and visuospatial working memory tasks, and verbal and visuospatial paired-associates learning tasks. In Experiment 1, older adults were generally slower than young adults on all speeded tasks, but age-related slowing was much more pronounced on visuospatial tasks than on verbal tasks. In Experiment 2, older adults showed smaller memory spans than young adults in general, but memory for locations showed a greater age difference than memory for letters. In Experiment 3, older adults had greater difficulty learning novel information than young adults overall, but older adults showed greater deficits learning visuospatial than verbal information. Taken together, the differential deficits observed on both speeded and unspeeded tasks strongly suggest that visuospatial cognition is generally more affected by aging than verbal cognition.

How General Is General Slowing? Evidence From the Lexical Domain

Three analyses are reported that are based on data from 19 studies using lexical tasks and a reduced version of the Hale, Myerson, and Wagstaff (1987) nonlexical data set. The results of Analysis 1 revealed that a linear function with a slope of approximately 1.5 described the relationship between the lexical decision latencies of older (65-75 years) and younger (19-29 years) adults. The results of Analysis 2, based on response latencies from 6 lexical tasks other than lexical decision, revealed a virtually identical linear relationship. In Analysis 3, it was found that performance on nonlexical tasks spanning the same range of task difficulty was described by a significantly steeper regression line with a slope of approximately 2.0. These findings suggest that although general cognitive slowing is observed in both domains, the degree of slowing is significantly greater in the nonlexical domain than in the lexical domain. In addition, these analyses demonstrate how the meta-analytic approach may be used to determine the limits to the external validity of experimental findings.

Selective Interference With the Maintenance of Location Information in Working Memory

In 3 experiments, the nature of the events that interfere with spatial working memory was examined in order to clarify the roles of imagery, attention, and other processes in the short-term maintenance of location information. Looking and pointing at secondary task stimuli selectively interfered with memory for the locations of primary task stimuli. Secondary tasks that involved either mentally rotating primary task stimuli or making color or shape discriminations about primary or secondary task stimuli interfered with spatial working memory only if the required response was visually guided, but not if the response was verbal. Taken together, these findings support P.S. Goldman-Rakics (1987) hypothesis regarding multiple representational domains and are consistent with known properties and connections of neurons believed to subserve the perception and maintenance of spatial information. A working memory task may be defined as one that requires holding onto information for a short time while it or other information is processed. Baddeley has proposed that there is a working memory system, specialized for concurrent storage and manipulation of information that is engaged by such tasks (for a recent overview, see Baddeley & Hitch, 1994). On the basis of studies of both intact and brain-damaged subjects (for a review, see Gathercole, 1994), Baddeley and his colleagues have divided the working memory system into three components: a phonological store for verbal information, a visuospatial sketchpad for visuospatial information, and a central executive that directs activities involved in manipulating and maintaining information in the two specialized stores. Perhaps the most studied example is the role of covert articulatory rehearsal in refreshing information in the phonological store. Performing a secondary task that requires articulation of verbal material decreases verbal memory span, a finding consistent with the hypothesized role of subvocalization in maintaining verbal information (e.g., Baddeley, Lewis, & Vallar, 1984). Much of the supporting evidence for the visuospatial component of Baddeleys model also comes from studies that use dual task procedures. These studies were designed to determine whether visuospatial and verbal working memory involve separate resources, and they have demonstrated that secondary verbal tasks interfere with memory for verbal information to a much greater extent than they interfere

Verbal and spatial working memory in school-age children : Developmental differences in susceptibility to interference

The development of verbal and spatial working memory was investigated with an interference paradigm. Memory spans were obtained from 3 groups (8-, 10-, and 19-year-olds) under 6 different conditions: Two primary memory tasks (1 verbal, 1 spatial) were administered in isolation and in conjunction with 2 versions of a secondary task. The primary tasks required recalling a series of visually presented digits and recalling the locations of Xs in a series of visually presented grids. The secondary tasks required reporting the color of the stimuli as they were presented using either a verbal or a spatial response. Analyses revealed that all age groups showed domain-specific interference (i.e., interference by a secondary task from the same domain as the primary task), but only the 8-year-olds also showed nonspecific interference (i.e., interference by a secondary task from a domain different than the primary memory task), suggesting that at least some executive functions do not reach adult levels of efficiency until approximately age 10.

Cognitive Processing Speed in Older Adults: Relationship with White Matter Integrity

Cognitive processing slows with age. We sought to determine the importance of white matter integrity, assessed by diffusion tensor imaging (DTI), at influencing cognitive processing speed among normal older adults, assessed using a novel battery of computerized, non-verbal, choice reaction time tasks. We studied 131 cognitively normal adults aged 55–87 using a cross-sectional design. Each participant underwent our test battery, as well as MRI with DTI. We carried out cross-subject comparisons using tract-based spatial statistics. As expected, reaction time slowed significantly with age. In diffuse areas of frontal and parietal white matter, especially the anterior corpus callosum, fractional anisotropy values correlated negatively with reaction time. The genu and body of the corpus callosum, superior longitudinal fasciculus, and inferior fronto-occipital fasciculus were among the areas most involved. This relationship was not explained by gray or white matter atrophy or by white matter lesion volume. In a statistical mediation analysis, loss of white matter integrity mediated the relationship between age and cognitive processing speed.

Effects of age, domain, and processing demands on memory span: Evidence for differential decline.

Analysis of cross-sectional data from the normative sample of the Wechsler Memory Scale - Third Edition (WMS-III) revealed different patterns of age-related differences in memory span measures depending on the type of memory item, processing demands, and the age of the older adult group. Regression of memory span on age revealed that the slope for Spatial Span raw scores was significantly more negative than the slope for Digit Span raw scores. There was no significant difference, however, either between the slopes for forward and backward Digit Span or between the slopes for forward and backward Spatial Span. Regression of Letter-Number Sequencing raw scores on age showed a distinctive, curvilinear pattern. Taken together, the present findings suggest that at least two mechanisms are involved in age-related differences in memory span. One mechanism, associated with a relatively linear decrease in memory span as a function of age, may differentially affect the storage of different types of information (e.g., sequences of digits vs. spatial locations). The other mechanism, evidenced by the curvilinear trend in Letter-Number Sequencing scores, may be tentatively attributed to a decline in executive aspects of working memory that becomes increasingly pronounced with age.