Perception and Reality of Cognitive Function: Information Processing Speed, Perceived Memory Function, and Perceived Task Difficulty in Older Adults

Participants

Community-dwelling older adults (n = 100) were recruited through adverts placed in local newspapers and social clubs throughout the Swansea area and by word of mouth. Exclusion criteria included poor self-reported general health, any past history of significant medical, neurological, or mental health problems, evidence of physical slowing (e.g., related to Parkinson’s disease or arthritis), or previous visit to a health care professional with memory complaints, anxiety, or depression. From those recruited, 19 individuals were excluded. Of these, 8 had Montreal Cognitive Assessment (MoCA) scores of 24 or 25, 3 had MoCA scores below this; 8 had current or past history of significant medical problems, anxiety, or depression. Of those included in the study (n¹ = 81; age 50+ y; 31 male, 50 female), all had normal or corrected-to-normal vision and hearing. Although medication could not be controlled, none of the participants reported receiving medication likely to affect information processing speed or cognitive function. All had normal overall cognition (score of 26 or above) using the MoCA [33] and no significant anxiety or depression, as determined by the Generalized Anxiety Disorder 7-item (GAD-7) score less than 5[34, 35] and the Patient Health Questionnaire (PHQ-9) score less than 4 [36], respectively. Participants’ age, sex, and years of full time education were recorded. A trained researcher administered the tests. Testing took place within the Psychology Department at the University of Swansea.

Subjective memory assessment

Subjective memory function was measured using the Memory Functioning Questionnaire (MFQ) [37]. This 64-item questionnaire assesses the perception of everyday memory functioning with seven sections on general rating of memory, retrospective functioning (compares current memory with past ability), frequency of forgetting, frequency of forgetting while reading, remembering past events, seriousness of forgetting (how memory impairment impacts daily life), and mnemonics usage. Each item is scored on a 1 to 7 Likert scale (1 = severe memory problems; 7 = no problems). Scores range between 64 and 448 with high scores reflecting less severe memory complaints.

Table 1 shows the demographics and TMT data.

Trail Making Test

Practice trails were provided for both TrailsA and B. For Trails A, the participants were instructed to draw one continuous line joining a series of circled numbers in ascending order on a sheet of paper as fast but as accurately as possible. For Trails B, the participants were instructed to draw one continuous line joining a series of circled numbers and letters alternately in ascending and alphabetical order on a sheet of paper as fast but as accurately as they could. Test outcome was the time taken in seconds to complete the test (with the time required to rectify any error forming part of the information processing speed score). No performance feedback was provided. Immediately after completing each of trails A and B, participants were asked to rate, using a scale of 1 to 7, how easy or difficult they found each test to complete, with 1 very easy to complete and 7 very difficult. Study debrief was performed at the end of the experimental session.

TMT: Information processing speed and subjective memory function

Spearman’s correlational analysis showed no association between MFQ total score and information processing speed for either Trails A or B (all p values > 0.05). Of the seven subscales of the MFQ, none showed an association with either Trails A or Trails B (all p values > 0.05) except for a significant negative correlation of Trails A with the Mnemonics subscale (r = –0.295, p = 0.007, which survives Bonferroni correction, p = 0.042).

TMT: Information processing speed and perceived task difficulty

The mean (sd) perceived task difficulty score was 2 (1.2) with a range from 1 to 6 for Trails A and 3 (1.6) with a range of 1 to 6 for Trails B. This was not significantly correlated with performance on Trails A (p > 0.05) but was significantly positively correlated with performance on Trails B (r = 0.293, p = 0.008), with slower actual information processing speed associated with a greater perceived task difficulty. Post hoc tests revealed that these results did not vary with respect to educational level or whether the participant was male or female.

Subjective memory function and perceived task difficulty

MFQ total score was significantly negatively correlated with perceived task difficulty for Trails A (r = –0.275, p = 0.013) and Trails B (r = –0.334, p = 0.002), with higher levels of subjective memory complaint related to greater perception of task difficulty. Post hoc tests revealed that these results did not vary with respect to educational level or whether the participant was male or female.

Participants

In the second study, another (separate) group of older adults (n = 62) were recruited. The protocol (i.e., inclusion and exclusion factors) was exactly the same as in study 1, as was the recruitment procedure. From those recruited, 6 individuals were excluded due to MOCA scores of 25 or less, with 2 further individuals excluded as a result of current poor medical health. Demographic details for the participants (n² = 54; age 50+ y; 24 male, 30 female) of this second study are shown in Table 2. All participantscompleted all 64 items on the MFQ.

Visual Search Task: Experimental task and procedure

For the computer-based visual search task, the time taken to respond to a target (target discrimination) when it appeared in isolation upon the screen and the time taken to respond to the same target when it was surrounded by similar but irrelevant and distracting stimuli were determined. This paradigm was presented on a Dell Precision PC running on Windows XP X86 CPU, viewed at a distance of 57 cm. All trials included a black target that was either a left- or right-pointing arrow, the task being to indicate whether the arrow was pointing to the right or left. The distracting stimuli consisted of seven black arrows that pointed up or down. A clock-face configuration (see Fig. 2) was used to position the target, both when it appeared alone and when surrounded by 7 distracters, in a counterbalanced arrangement in order to eliminate any differences in processing between right and left and upper and lower visual fields. A total of 64 trials were presented; the target appearing 8 times at each of the possible ‘clock-face’ locations. For half of the trials distracters were presented at the other locations and for the other half no distracters were presented. For each trial the central fixation cross appeared on screen for 1000 ms prior to the appearance of the target (with or without distracters) and remained on screen for the duration of the trial. The stimuli remained on screen until the participant responded, after which the fixation point re-appeared. The participants were instructed to fixate on the center cross at the beginning of each trial and to respond as quickly but as accurately as possible to whether the target was pointing to the right or left by pressing one of two computer keyboard keys. After instruction, all participants were asked to reiterate the instructions to ensure understanding and then performed a practice block of no more than 10 trials. The ability of the participants to fixate on the cross at the beginning of each trial continued to be checked throughout the procedure by researcher observation. No performance feedback was provided.

Fig.1

Box plot of mean information processing speed (s) for Trails A and B performance in older adults.

Fig.2

Search stimulus.

Group mean errors were calculated. Responses were eliminated if they were incorrect or obviously due to a disturbance/lapse of concentration or below 150 ms (faster than ‘natural’ reaction time therefore representing the pre-empting of the stimulus). No participants failed to respond to a trial. For each participant, the median time (information processing speed) taken to respond for the target alone and the target plus distracter trials was determined and group mean data produced (seeTable 2).

Perception of task difficulty

Immediately after completing the test participants were asked to rate, using a Likert scale of 1 to 7, how easy or difficult they found each test to complete, with 1 very easy to complete and 7 very difficult. Study debrief was performed at the end of the experimental session.

Visual search: information processing speed and subjective memory function

MFQ total score and subscales scores were not significantly correlated with information processing speed for either the target alone or the target plus distracters conditions (p values > 0.05). Overall errors on the visual search tasks were very small (mean group errors 0.37 for target alone and 0.33 for target plus distracters) and the number of errors was not significantly correlated with MFQ scores (p values > 0.05).

Visual search: information processing speed and perceived task difficulty

The mean (sd) perceived task difficulty score for the visual search task was 3 (1.4) with a range from 1 to 6. This was significantly negatively correlated with the target alone condition (r = –0.294, p = 0.031), with slower actual information processing speed associated with less perceived task difficulty, but not for the target plus distracters conditions (p > 0.05). The number of errors was not significantly correlated with perceived task difficulty (p values > 0.05).

Subjective memory function and perceived task difficulty

MFQ total score was not significantly correlated with perceived task difficulty for either the target alone or the target plus distracters condition (p values > 0.05). Post hoc tests revealed that all the above results did not vary with respect to whether theparticipant was male or female.

Educational level and information processing speed

Further post hoc analysis revealed that although educational level was not significantly related to subjective memory performance it was significantly positively correlated with perceived test difficulty (r = 0.440, p = 0.01) and it was significantly negatively correlated with information processing speed for target plus distracter (r = –0.398, p = 0.003) condition, but not for the target alone condition (p > 0.05), i.e., a higher level of education is related to faster information processing speed when distractors are present.

Information processing speed and subjective memory

Greater levels of overall subjective memory complaint (MFQ total score) were not significantly associated with slower information processing speed as measured by the TMT or Visual Search Task. For Trails A there was a significant negative correlation between information processing speed and the mnemonics subset of the MFQ and although this survived Bonferroni correction, the significance level of this effect was low (p = 0.042). For Trails B and both conditions of the visual search test, information processing speed was not significantly correlated with any of the MFQ subsets.

Given the close relationship between information processing speed and white and grey matter structure [6–12], the general absence of an association with perceived memory dysfunction in otherwise cognitively healthy, euthymic older adults suggests perceived memory dysfunction is less likely to be related to structural abnormality or possible neurodegenerative change. Similarly, given the association between Alzheimer’s disease, vascular dementia and MCI and slowed information processing speed the lack of a relationship between perceived memory function and information processing speed may be indicative of a non-neurodegenerative basis for perceived impairment in individuals with ‘normal’ levels of general cognition (e.g., MoCA score within the normal range).

Note, however, that these ideas are speculative in nature given the absence of neuroimaging, a full range of objective and subjective measures of memory, cognitive and information processing speed performance, and longitudinal analysis examining the risk of developing MCI and/or dementia. The possibilitythat changes in brain structure and function and in memory may occur in the absence of changes in information processing speed must also be considered. Individuals also may perceive problems with memory but still perform at normal information processing speed as there may be, for some tasks, factors which influence memory and perceived memory but not information processing speed.

Nevertheless, if such findings were found to be robust after further research and development, the measurement of behavioral information processing speed might be of use in helping to determine for whom priority should be given with respect to further investigation and follow up of subjective memory complaints. For example, disproportionately slower responses (such as the outliers evident in our results, see Figs. 1 and 3) may be representative of detrimental change in underlying structure and function and thus a greater possibility of underlying neurodegeneration. Clinical follow-up and medical intervention with these patients could be prioritized, whereas others may benefit more from focus on psychological support, adjusting expectations regarding normal changes in cognitive performance, and providing metacognitive strategies to reduce the required effort for everyday tasks.

Fig.3

Box plot of mean information processing speed (ms) for target alone and target plus distracters trials.

Perception of task difficulty and subjective memory

Greater subjective memory complaint was associated with a perception of greater task difficulty for both Trails A and Trails B of the TMT, especially for the more difficult Trails B test, with higher levels of subjective memory complaint related to greater perceived task difficulty. With respect to the TMT therefore, individuals who reported higher levels of perceived memory dysfunction (total MFQ score) are those who also reported the greater levels of task difficulty, despite actual (objectively measured) information processing speed not being associated with perceived memory function. To speculate, the relationship between perceived memory and perceived task difficulty in individuals with MOCA scores within the normal range in the absence of objective change in information processing speed suggests that perceived memory dysfunction may be related more to metacognitive factors than underlying structural change.

In contrast, for the visual search task, greater subjective memory complaint was not associated with a perception of greater task difficulty for both the target alone and the target plus distracter conditions. Individuals who reported higher levels of memory dysfunction did not report greater levels of taskdifficulty.

Such outcome variability indicates that the metacognition relationship between perceived memory and difficulty of task may not be generalizable to all tests, but mediated by factors such as difficulty, resource requirements and brain functions and areas recruited. However, different participants were used in the two studies and so the outcome effects may not be directly comparable. Nevertheless, our results raise the possibility that examining patterns of perceived memory function and the perception of task difficulty may help to determine whether perceived memory impairment (in the absence of objective change in memory function) is related to structural change or metacognition (which is more likely to be responsive to intervention and treatment than structural change), or indeed whether a much more complex relationship exists between metacognition, structural change and actual and perceived functional integrity.

Information processing speed and perceived task difficulty

For Trails A, perceived task difficulty was not significantly correlated with objectively measured information processing speed. In contrast, perceived task difficulty was positively correlated with information processing speed for Trails B, i.e., slower information processing speed was associated with greater perceived task difficulty. For the target plus distracters condition of the visual search task, perceived task difficulty was not significantly correlated with objectively measured information processing speed. For the target alone condition however, perceived task difficulty was significantly negatively correlated with information processing speed, i.e., slower information speed was associated with a lower level of perceived task difficulty. This pattern of results indicates that, irrespective of perceived memory function, the judgement of task difficulty is related to the nature of the task and is not always related to actual performance.

Educational level

For both Trails A and B, educational level was not significantly associated with information processing speed, subjective memory function or perceived task difficulty. For the visual search task, educational level was also not significantly related to subjective memory performance but it was significantly positively correlated with perceived test difficulty, i.e., a higher level of education was associated with higher levels of perceived task difficulty. Furthermore, educational level was significantly negatively correlated with information processing speed for the target plus distracter condition, i.e., a higher level of education was associated with faster information processing. In contrast, educational level was not significantly correlated with information processing speed for the target alone condition. Although educational level was not significantly related to perceived memory function, this pattern of results indicates the potential influence of education upon perceived task performance and actual task performance per se and a failure to take such factors into account in research may influence the interpretation of study outcome.

Study limitations

The participants in each study cohort were relatively young (mean age 65 y). However, this is the age group when subjective memory changes are typically reported and when pathological changes of neurodegenerative disease start to become more common. Future work, including neuroimaging, needs to explore the structural and functional relationships between information processing speed, actual and perceived memory function and metacognition factors in older age groups and in those seeking clinical diagnosis and with respect to mood and personality. Further research is also required to investigate why some individuals do not approach health care services about their perceived memory function. Some individuals may be anxious about a formal diagnosis, fearful that they may be developing dementia and unaware of other possible reversible causes of impairment or potential interventions to improve quality of life.

We tested participants on only one occasion and could not take into account the possibility that perceived memory may fluctuate, as it can be due to temporary conditions such as fatigue, or everyday stressful life events. Perceived memory was measured using only the MFQ. Related to this issue is the fact that we were not able to determine the accuracy of self-report on this measure. Furthermore, we did not include a battery of tests objectively measuring memory and cognition, while information processing speed was measured by only two tests, both of which had large executive function components. Nor did we measure separate specific components of behavioral information processing speed, such as motor function. The relationship between actual and perceived memory function, information processing speed and indeed metacognitive processes such as the perception of task difficulty, remains to be determined in relation to a variety of different tests, and ones that recruit different brain structure and functional domains and specific metacognitive factors. Finally, although it may not be possible to control medication, future studies should record and report medication in order to facilitate a greater understanding the generalizability of results to the population in general.