Web-Based Cognitive Testing of Older Adults in Person Versus at Home: Within-Subjects Comparison Study.

BACKGROUND: Web-based research allows cognitive psychologists to collect high-quality data from a diverse pool of participants with fewer resources. However, web-based testing presents unique challenges for researchers and clinicians working with aging populations. Older adults may be less familiar with computer usage than their younger peers, leading to differences in performance when completing web-based tasks in their home versus in the laboratory under the supervision of an experimenter. OBJECTIVE: This study aimed to use a within-subjects design to compare the performance of healthy older adults on computerized cognitive tasks completed at home and in the laboratory. Familiarity and attitudes surrounding computer use were also examined. METHODS: In total, 32 community-dwelling healthy adults aged above 65 years completed computerized versions of the word-color Stroop task, paired associates learning, and verbal and matrix reasoning in 2 testing environments: at home (unsupervised) and in the laboratory (supervised). The paper-and-pencil neuropsychological versions of these tasks were also administered, along with questionnaires examining computer attitudes and familiarity. The order of testing environments was counterbalanced across participants. RESULTS: Analyses of variance conducted on scores from the computerized cognitive tasks revealed no significant effect of the testing environment and no correlation with computer familiarity or attitudes. These null effects were confirmed with follow-up Bayesian analyses. Moreover, performance on the computerized tasks correlated positively with performance on their paper-and-pencil equivalents. CONCLUSIONS: Our findings show comparable performance on computerized cognitive tasks in at-home and laboratory testing environments. These findings have implications for researchers and clinicians wishing to harness web-based testing to collect meaningful data from older adult populations.

Effects of Question Framing on Self-Reported Memory Concerns across the Lifespan.

Background/Study Context: Evidence regarding whether there is an age-related increase in subjective memory concerns is mixed. The goal of this study was to investigate whether calling to mind specific instances of memory failures affects the likelihood that individuals report being concerned about their memory. METHODS: Young, middle-aged, and older individuals responded to statements that probed general memory concerns (e.g., I am concerned about my memory) before or after probing the frequency of memory difficulties in specific everyday situations (e.g., I forget appointments). RESULTS: We found no relationship between age and memory concern, and older compared to younger adults reported having fewer everyday memory failures. Furthermore, individuals of all ages were more likely to report being concerned about their memory if asked before relative to after rating the frequency of specific everyday memory failures. CONCLUSION: Increasing age is associated with fewer reported difficulties with specific everyday memory situations, but people of all ages display anchoring effects on memory concern based on realistic occurrences of remembering. Our findings have implications for the measurement of self-reported memory as they show that the order of questions can influence responding.

Learning from your mistakes: does it matter if you're out in left foot, I mean field?

Studies have shown that generating errors prior to studying information (pencil-?) can improve target retention relative to passive (i.e., errorless) study, provided that cues and targets are semantically related (pencil-ink) and not unrelated (pencil-frog). In two experiments, we manipulated semantic proximity of errors to targets during trial-and-error to examine whether it would modulate this error generation benefit. In Experiment 1, participants were shown a cue (band-?) and asked to generate a related word (e.g., drum). Critically, they were given a target that either matched the semantic meaning of their guess (guitar) or mismatched it (rubber). In Experiment 2, participants studied Spanish words where the English translation either matched their expectations (pariente-relative) or mismatched it (carpeta-folder). Both experiments show that errors benefit memory to the extent that they overlap semantically with targets. Results are discussed in terms of the retrieval benefits of activating related concepts during learning.

The benefits of errorless learning for people with amnestic mild cognitive impairment.

The aim of this study was to explore whether errorless learning leads to better outcomes than errorful learning in people with amnestic mild cognitive impairment (MCI), and to examine whether accuracy in error recognition relates to any observed benefit of errorless over errorful learning. Nineteen participants with a clinical diagnosis of amnestic MCI were recruited. A word-list learning task was used and learning was assessed by free recall, cued recall and recognition tasks. Errorless learning was significantly superior to errorful learning for both free recall and cued recall. The benefits of errorless learning were less marked in participants with better error recognition ability. Errorless learning methods are likely to prove more effective than errorful methods for those people with MCI whose ability to monitor and detect their own errors is impaired.

Mistakes as stepping stones: Effects of errors on episodic memory among younger and older adults.

The memorial costs and benefits of trial-and-error learning have clear pedagogical implications for students, and increasing evidence shows that generating errors during episodic learning can improve memory among younger adults. Conversely, the aging literature has found that errors impair memory among healthy older adults and has advocated for the use of errorless learning to rehabilitate memory. However, there is evidence that errors are not always beneficial for younger adults, nor always harmful for older adults. We propose that differences in the learning paradigms used in the younger and older adult literatures may account for these conflicting recommendations, namely that they typically engender conceptual and nonconceptual processing, respectively. In this study, we had younger and older adults study words under errorless and trial-and-error learning instructions and based either on conceptual (a flower--tulip) or lexical (ho___--house) cues. We found that relative to errorless learning, trial-and-error learning increased target memory in the conceptual condition but decreased it in the lexical condition. Critically, both age groups showed this pattern, implying that aging does not influence how we learn from mistakes. We suggest that conceptual guesses act as "stepping stones" toward the target whereas lexical guesses simply create retrieval noise. This suggestion was supported by the fact that participants of both ages remembered their prior guesses better in the conceptual than lexical condition and that memory for guesses mediated differences in target performance. These findings are discussed within the framework of current theories on the effects of error generation on episodic memory.

Updating misconceptions: effects of age and confidence.

Young adults are more likely to correct an initial higher confidence error than a lower confidence error (Butterfield & Metcalfe, 2001). This hypercorrection effect has never been investigated among older adults, although features of the standard paradigm (free recall, metacognitive judgments) and prior evidence of age-related error resolution deficits (see Clare & Jones, 2008) suggest that they may not show this effect. In Study 1, we used free recall and a 7-point confidence scale; in Study 2, we used multiple-choice questions, and participants indicated how many alternatives they had narrowed their options down to prior to answering. In both studies, younger and older adults showed a hypercorrection effect, and this effect was equivalent between groups in Study 2 when free recall and explicit confidence ratings were not required. These results are consistent with our previous work (Cyr & Anderson, 2012) showing that older adults can successfully resolve learning errors when the learning context provides sufficient support.

Contributions of frontal and medial temporal lobe functioning to the errorless learning advantage.

Among individuals with episodic memory impairments, trial-and-error learning is less successful than when errors are avoided. This "errorless learning advantage" has been replicated numerous times, but its neurocognitive mechanism is uncertain, with existing evidence pointing to both medial temporal lobe (MTL) and frontal lobe (FL) involvement. To test the relative contribution of MTL and FL functioning to the errorless learning advantage, 51 healthy older adults were pre-experimentally assigned to one of four groups based on their neuropsychological test performance: Low MTL-Low FL, Low MTL-High FL, High MTL-Low FL, High MTL-High FL. Participants learned two word lists under errorless learning conditions, and two word lists under errorful learning conditions, and memory was tested via free recall, cued recall, and source recognition. Performance on all three tests was better for those with High relative to Low MTL functioning. An errorless learning advantage was found in free and cued recall, in cued recall marginally more so for those with Low than High MTL functioning. Participants with Low MTL functioning were also more likely to misclassify learning errors as target words. Overall, these results are consistent with a MTL locus of the errorless learning advantage. The results are discussed in terms of the multi-componential nature of neuropsychological tests and the impact of demographic and mood variables on cognitive functioning.

Trial-and-error learning improves source memory among young and older adults.

Trial-and-error learning, relative to errorless learning, has been shown to impair memory among older adults, despite evidence from young adults that errors may afford memorial benefits through richer encoding. However, previous studies on the effects of errorless versus trial-and-error learning in older adults has required production of errors based on perceptual cues. We hypothesized that producing errors conceptually associated with targets would boost memory for the encoding context in which information was studied, especially for older adults who do not spontaneously elaborate on targets at encoding. We report two studies examining the impact of generating errors during learning on source memory among young and older adults, with a process dissociation procedure employed in Study 1, and source memory assessed directly in Study 2. In both studies, participants were shown semantic category cues and generated an exemplar either with or without errors. In Study 1, for both age groups trial-and-error learning was associated with lower familiarity-based memory and higher recollection-based memory relative to errorless learning, and the latter effect was more marked for older than younger adults. Similarly, in Study 2, trial-and-error learning was associated with better source memory relative to errorless learning, particularly for the older adults. We argue that trial-and-error learning can enhance source memory and confer memorial benefits when making such errors facilitates semantic elaboration, especially for older adults who do not spontaneously engage in strategic encoding.

Driving difficulties of brain-injured drivers in reaction to high-crash-risk simulated road events: a question of impaired divided attention?

This study examined the role of impaired divided attention and speed of processing in traumatic brain injury (TBI) drivers in high-crash-risk simulated road events. A total of 17 TBI drivers and 16 healthy participants were exposed to four challenging simulated roadway events to which behavioral reactions were recorded. Participants were also asked to perform a dual task during portions of the driving task, and TBI individuals were administered standard measures of divided attention and reaction time. Results indicated that the TBI group crashed significantly more than controls (p < .05) and that dual-task performance correlated significantly with crash rate (r = .58, p = .05).