Journal of Natural Science, Biology and Medicine

ORIGINAL ARTICLE
Year
: 2021  |  Volume : 12  |  Issue : 1  |  Page : 97--102

“I Think I Can Remember” age-related changes in self-efficacy for short-term memory


Dasmine Fraclita DSouza1, Gagan Bajaj1, Vinitha Mary George2, Sudhin Karuppali1, Jayashree S Bhat1,  
1 Department of Audiology and Speech Language Pathology, Kasturba Medical College, Manipal Academy of Higher Education, Mangalore, Karnataka, India
2 Department of Audiology and Speech Language Pathology, National Institute of Speech and Hearing, Thiruvananthapuram, Kerala, India

Correspondence Address:
Gagan Bajaj
Department of Audiology and Speech Language Pathology, Kasturba Medical College, Manipal Academy of Higher Education, Mangalore, Karnataka
India

Abstract

Introduction: Changes in metacognitive abilities due to aging, like self-efficacy, have received less attention in cognitive research. Short-term memory (STM) declines among aging adults are well known but the age-related trends of self-efficacy linked to the same have received less attention. The present research aimed at studying age-related trends in self-efficacy linked to STM among the young-aged, middle-aged, and old-aged adults. Materials and Methods: Participants performed face recall, name recall, object recall, face-name association, first-second name association, and face-object association tasks. The self-efficacy linked to these STM tasks was measured through a pre-task prediction question and a post-task judgment question. Descriptive statistics and two-way mixed model ANOVA with post hoc Bonferroni analysis were performed to assess age related changes in self-efficacy measures. Results: The findings revealed significant overestimation of performance, during pretask prediction, by old-aged adults and middle-aged adults. While the posttask judgment was recalibrated closer to the actual performance by participants of all age groups. Conclusion: The current research findings indicate that self-efficacy for STM follows an age related decline. Therefore, inclusion of self-efficacy measures in the assessment of STM would provide a valuable insight as it describes an individual's own awareness about their STM abilities, provides realistic feedback about one's STM performance and also aids clinicians in understanding the perception-performance dynamics among the aging adults.



How to cite this article:
DSouza DF, Bajaj G, George VM, Karuppali S, Bhat JS. “I Think I Can Remember” age-related changes in self-efficacy for short-term memory.J Nat Sc Biol Med 2021;12:97-102


How to cite this URL:
DSouza DF, Bajaj G, George VM, Karuppali S, Bhat JS. “I Think I Can Remember” age-related changes in self-efficacy for short-term memory. J Nat Sc Biol Med [serial online] 2021 [cited 2021 Apr 13 ];12:97-102
Available from: http://www.jnsbm.org/text.asp?2021/12/1/97/307854


Full Text



 Introduction



Short-term memory (STM), a cognitive mechanism is defined as the “faculties of the human mind that can hold a limited amount of information in a very accessible state temporarily”.[1] Retrieval of information from STM can be in the form of recall referring to the constant re-access of information and association referring to the retrieval by linking two or more items, from a previously presented original set of items.[2],[3] STM is essential for higher order actions such as troubleshooting, developing plans, and problem solving.[4] It includes recalling faces, objects, names of people and making associations between first and second names of a person, names of people, their faces, the objects they possess and age-related STM decline have been reported to begin from middle twenties to old age.[5],[6] The awareness of decline in STM abilities is critical to understand the real capacities in deciding how one utilize recollections in ordinary circumstances.[7] Self-efficacy is defined as “people's beliefs about their capabilities to produce designated levels of performance that exercise influence over events that affect their lives.”[8] Self-efficacy beliefs are responsible for the regulation of an individual's feelings, thoughts, motivation, and behaviors. Individuals with better awareness about their memory abilities, are able to implement appropriate strategies, thus take maximal advantage of their memory.[9] The existing research regarding effect of age on self-efficacy linked to memory has given mixed understanding. For example, there are some reports which suggest an overestimation of performance on memory tasks by aging adults[10],[11] and some research suggests that aging adults possess relatively accurate predictions on memory tasks.[12],[13] Considering the impact which STM deficits can pose on an aging individual, it would be of great importance for the cognitive researchers and clinicians to apprehend the age related trends in the self-efficacy associated to STM abilities. Self-efficacy abilities linked to STM have been assessed in the past using questionnaire-based method,[14] pretask prediction paradigm, and postdiction method.[15] The prediction-postdiction method may offer better insight into one's self-efficacy linked to memory functions as it is more specific to the given task and stimuli than general questionnaires.[16] An individual's awareness about their memory changes may ultimately drive their primary prevention, early identification, and timely intervention, if required. Hence, the present research aimed at studying the changes in self-efficacy in relation to STM performance among the young-aged adults, middle-aged adults, and old-aged adults.

 Materials and Methods



Study design

The study followed a cross-sectional research design. The institutional ethical board had approved the research protocol.

Participants

A total of 66 participants, with 22 participants (males = 11 and females = 11) from each of the three groups, i.e., the young-aged adults (Mean = 27.5 years, standard deviation [SD] = 6.5 years), middle-aged adults (M = 50.7 years, SD = 4.2 years), and old-aged adults (M = 69.6 years, SD = 4.1 years) were recruited. All the participants belonged to the middle socioeconomic status on the basis of the national socioeconomic status scale for urban population[17] and possessed a minimum of 15 years of formal education in an English medium educational environment. Statistically significant difference was not found with regard to number of years of formal education between the participants of the three aged groups (F(2,65) = 1.891, P = 0.159). Based on the National Institute on Alcohol Abuse and Alcoholism scale[18] and the classification of smoking,[19] only participants who were nondrinkers and nonsmokers respectively were included in the study. Mini-mental state examination (MMSE)[20] was performed on the participants for ruling out the presence of occult cognitive impairment. Based on the Indian norms[21] of MMSE, a cutoff score of 26 was considered to recruit the participants. Demographic data and MMSE scores of the three groups are shown in [Table 1]. An informed consent was obtained from all the participants prior to their recruitment in this research.{Table 1}

Materials and stimuli

The assessment of everyday STM comprised of six tasks, which included face recall, name recall, object recall, face-name association, first-second name association, and face-object association tasks. All the STM tasks were prepared in the visual modality. A set of 100 bisyllabic Indian names with an equal representation of gender and community were shortlisted for the STM tasks involving names as stimuli. For the STM tasks involving faces as stimuli, photographs of young Indian adult faces were taken. Prior written consent was taken from each participant before including their photograph as the stimulus. Measures were taken during photography to maintain uniformity across the participants with respect to their clothing, hairstyle, ornaments, seating, lighting, background, and facial expressions. In this manner, a total of 67 male faces and 84 female faces photograph were taken. For STM tasks involving objects as the stimuli, a list comprising of 150 most commonly occurring objects in Indian context was prepared. All these 150 common objects were found to have high lexical frequency as rated by 3 experienced speech language pathologists on a 3-point rating scale where “3” represented high lexical frequency and “1” represented low lexical frequency. The photographs of each of these real objects were taken with a common background, appropriate distance, and optimum brightness. These stimuli were then validated for their technical suitability and appropriateness for the study by three speech language pathologists with more than 5 years of experience in the field of cognitive communicative research. The validated stimuli were further arranged in increasing order in series, ranging from 2 to 7 stimuli per series, with no repetition of stimulus within and across tasks. Each complexity level consisted of one trial. Within a series of stimuli, the recall tasks comprised of one stimulus per presentation, whereas the association tasks comprised of two stimuli per presentation, for example, a face and an object for face-object association, two names for first and second name association and a face and a name for face-name association. The experiments were embedded in the licensed version of Paradigm software 2.5.0.68. The duration of stimulus presentation was 2000 ms and inter-stimulus duration was 1000 ms. [Figure 1] depicts an example of a 2-step face-object association task. [Table 2] shows the technical details of STM tasks used in the present research. The experimental tasks were presented visually on a 15-inch laptop screens. The participants were asked to verbally recall the target items across all the tasks and these verbal responses were recorded through a microphone.{Figure 1}{Table 2}

Self-efficacy in both pretask and posttask condition was assessed with the pre/post-diction questions.[11] The questions for the pretask prediction and posttask judgment are shown in [Table 3]. The pretask predictions and posttask judgments made by the participants were noted by the researcher in the response sheet which was not accessible to the participant for any later reference.{Table 3}

Research procedure

For each STM task, the pretask prediction question was asked soon after the practice trials, followed by the task performance. The experiment was terminated when the participants performed incorrectly in two consecutive steps, and the STM span was considered to be the last level or step where the participants could perform accurately. Immediately after completing the task, the posttask judgment question was asked.

Statistical analysis

Two-way mixed model ANOVA with group as a between-subject factor and time as a repeated factor was performed. This was followed by post hoc Bonferroni pairwise comparisons to seek the relation of self-efficacy with the actual performance on the STM tasks across the three age groups. SPSS 16.0 version was used. Significance value of <0.05 was considered on statistical tests.

 Results



All age groups overestimated the prerecall prediction when compared to the actual task, whereas the postrecall judgment was found to be similar or overlapping with actual recall performance. The group-wise average of prerecall prediction, recall, and postrecall judgment scores across the three STM recall tasks are summerized in [Figure 2]. The average prerecall prediction scores suggest that young-aged adults (M = 5.1, SD = 1.2) predicted their recall abilities higher than the middle-aged adults (M = 4.7, SD = 1.2) and old-aged adults (M = 4.4, SD = 1.6), respectively. However, the extent of overestimation in pretask condition was larger in the middle- and old-aged adults than the young aged adults. Similar pattern was observed for average post recall judgment scores (young-aged adults - M = 4.5, SD = 0.6; middle-aged adults - M = 4.1, SD = 0.8; old-aged adults - M = 3.2, SD = 0.8). The middle-aged adults were closest to their actual performance as compared to young- and old-aged adults who tend to slightly underestimate their performance.{Figure 2}

[Figure 3] shows similar trends for the STM association tasks, where the average preassociation absolute prediction scores were greater for young aged adults (M = 4.2, SD = 1.1) than middle aged (M = 4.11, SD = 1.1) and old aged adults (M = 3.6, SD = 0.9). Similar pattern was observed for the average post association judgment scores (young-aged adults - M = 3.6, SD = 0.5; middle-aged adults - M = 3.5, SD = 0.7; old-aged adults - M = 2.7, SD = 0.7). During pretask prediction, the middle-aged and old-aged adults overestimated to a larger extent in comparison with young aged adults. In post task judgment, middle-aged adults were closest to their actual performance in comparison to young- and old-aged adults who underestimated their post task judgment with respect to their actual performance.{Figure 3}

A two-way mixed model ANOVA with group as a between-subject factor and time (three measurements, i.e., prerecall prediction, recall, postrecall judgment) as a repeated factor was performed. The results did not show any group * time interaction of self-efficacy for any of the recall (face recall - F(2,126) = 0.808, P = 0.485; name recall - F(2,126) = 0.703, P = 0.591 and object recall - F(2,126) = 0.943, P = 0.441) and association tasks (name association - F(2,126) = 1.483, P = 0.211; face name association - F(2,126) = 0.168, P = 0.954 and face object association - F(2,126) = 1.209, P = 0.310) respectively.

The effect of time was significant for two recall tasks (face recall - F(2,126) = 20.092, P < 0.001; name recall - F(2,126) = 22.487, P < 0.001) and all association tasks (name association - F(2,126) = 17.633, P < 0.001; face name association - F(2,126) = 18.308, P < 0.001; and face object association = F (2,126) =4.609, P = 0.012). The effect of time was not significant for the object recall task (object recall - F(2,126) = 1.876, P = 0.157).

The effect of group was observed for all the recall tasks (face recall - F(2,63) = 19.379, P < 0.001; name recall - F(2,63) = 11.903, P < 0.001 and object recall - F(2,63) = 6.097, P = 0.004) and association tasks except for the name association task (name association - F(2,63) = 2.124, P = 0.128; face name association - F(2,63) = 8.211, P = 0.001; and face object association - F(2,63) = 12.126, P < 0.001). These results are shown in [Table 4].{Table 4}

Following which, Bonferroni post hoc test was applied and pairwise comparisons were performed to understand the main effects observed for time and group, respectively. The results of time effects, as shown in [Table 5], suggest that participants significantly overestimated their STM performance for the face recall (P < 0.001), name recall (P < 0.001), name association (P < 0.001), and the face name association (P < 0.001). The group effects suggested that the overestimation significantly increased with age for face recall, name recall, and face name association tasks, i.e., old-aged adults significantly overestimated their performance prior to the task as compared to young-aged adults (face recall - P < 0.001, name recall - P = 0.000, face name association - P < 0.001) and middle-aged adults (face recall - P = 0.009, name recall - P = 0.027, face name association - P = 0.003). The overestimation by middle aged adults was significant as compared to the young-aged adults only for the name recall task (P = 0.011).{Table 5}

During posttask judgment, a statistically significant underestimation was observed only for the object recall task (P = 0.040). The group specific results indicated that older adults underestimated their performance significantly as compared to the middle aged adults (P = 0.031) and young aged adults (P = 0.001) for the object recall task. All other tasks were found to have statistically no significant difference as compared to the actual performance for any of the three groups (P > 0.05).

 Discussion



Overall performance in self-efficacy showed that the pretask prediction was higher than the actual task performance for the STM tasks in all age groups and more significantly in the middle and old aged adults. Looking at the magnitude of the overestimation, it appears that as an individual age, there is a tendency to significantly overestimate one's STM abilities. These findings fetch their support from previous literature[11],[22] who have reported an increased overestimation trend with age during pretask memory prediction.

Several reasons have been discussed in literature about aging adults overestimating their memory performance before the task. One viewpoint deals with sub vocal rehearsals which are carried out by the participants before performing the actual memory task. Literature reports that old aged adults experience difficulty with subvocal rehearsals before a memory task where >4 items have to be retained.[23]

People use inferential strategies while predicting their performance, called as “feeling of knowing.”[23] “Feeling of knowing” could be due to familiarity[24] which can be attributed to the trial given before the commencement of task hence leading in overestimation in prediction. Another possible explanation for overestimation of STM abilities among aging adults could be that old-aged adults may actually be underestimating the difficulty of the memory task.[25] Old-aged adults may not have tackled a similar memory task in their recent past and thus assume the STM task to be easier.[25]

Finally, there is usually a significant age related decline in the STM abilities. There is a good possibility that aging adults might be unaware of their age-related declines in STM. This unawareness of the memory related declines among aging adults can be described with reference to the literature.[26] There could primarily be four reasons that an individual is unaware of his/her age linked declines in cognition. First, there is seldom a need for an individual to perform at a maximum level in everyday life situations whereas the cognitive test assesses the maximum level of functioning of an individual. Second, there is a shift from unique processing to dependence on gathered knowledge as age progresses. This compensation might prevent an aging adult from manifesting his cognitive deficit on his everyday life. Third, it has been argued that one's performance in everyday life does not solely depend on the cognitive abilities of an individual. Fourth possible explanation for a mismatch between the performance and perception could be that aging adults learn to either minimize their exposure to deficit revealing situations or they do a selective optimization when they have to uphold a high level of expertise.

The posttask judgment for all the age groups was close to their actual performance. Similar trends for posttask judgment have been reported, who found that the accuracy of posttask judgment for aging adults was more precise and approximated closer to the actual performance than pretask predictions.[15] This suggests that individuals of all the age groups monitored their performance during the study-test cycle and recalibrated their self-efficacy. The recalibration abilities during the posttask judgment seems to be unaffected by age. This means that there is an age related sparing of the self-monitoring abilities. These spared self-monitoring abilities among the aging adults can be promising in a way that metacognitive training can be offered to old aged adults for improving their self-regulation abilities.[27]

The findings of the current research were obtained from a relatively smaller sample size and across limited STM tasks. It would be promising to generate appropriate evidence in this regard on a larger research group across wide range of STM tasks such as digits, letters, and other stimuli-related to actions of everyday life. Future research could also assess the effect of other age related personal factors such as education, employment, stress levels, and health conditions on the self-efficacy and STM dynamics. It would also be worthwhile to explore the relation of self-efficacy with STM abilities among individuals with neurogenic communication disorders such as aphasia, dementia, and traumatic brain injury.

 Conclusion



Outcomes of the current research suggest a significant over estimation of the STM abilities by middle- and old-aged adults during pretask prediction. While performing the posttask judgment, the participants of all the three age groups tend to recalibrate their self-efficacy according to their actual performance. This age-related over-estimation of pretask prediction and spared self-monitoring during posttask judgment could offer significant insight with respect to age linked changes in cognitive and linguistic processing. The study advocates the usage of self-efficacy measures in cognitive communicative assessment and treatment modules to promote healthy perception-performance dynamics among healthy and pathologically aging individuals.

Financial support and sponsorship

This work was supported by the Technology Interventions for Disabled and Elderly, Department of Science and Technology, Government of India (DST-TIDE) (Grant No. SEED/TIDE/005/2013-[C]).

Conflicts of interest

There is no conflicts of interest.

References

1Cowan N. What are the differences between long-term, short-term, and working memory? Prog Brain Res 2008;169:323-38.
2Mridula J, George VM, Bajaj G, Namratha HG, Bhat JS. Effect of working memory training on cognitive communicative abilities among young- and middle-aged adults. Cogent Psychol 2017;4.
3Namratha HG, George VM, Bajaj G, Mridula J, Bhat JS. Effect of yoga and working memory training on cognitive communicative abilities among middle aged adults. Complement Ther Clin Pract 2017;28:92-100.
4Shallice T. Specific impairments of planning. Philos Trans R Soc Lond B Biol Sci 1982;298:199-209.
5Crook TH, West RL. Name recall performance across the adult life-span. Br J Psychol 1990;81(Pt 3):335-49.
6Kail R, Salthouse TA. Processing speed as a mental capacity. Acta Psychol (Amst) 1994;86:199-225.
7McDougall GJ, Kang J. Memory self-efficacy and memory performance in older males. Int J Mens Health 2003;2:131-47.
8Bandura A. Self-efficacy. In: Ramachaudran VS, editor. Encyclopedia of Human Behaviour. New York: Academic Press; 1994. p. 71-81.
9Bertrand JM, Moulin CJ, Souchay C. Short-term memory predictions across the lifespan: Monitoring span before and after conducting a task. Memory 2017;25:607-18.
10Bunnell JK, Baken DM, Richards-Ward LA. The effect of age on metamemory for working memory. NZ J Psychol 1999;28:23-9.
11Touron DR, Oransky N, Meier ME, Hines JC. Metacognitive monitoring and strategic behaviour in working memory performance. Q J Exp Psychol (Hove) 2010;63:1533-51.
12Camp CJ, Markley RP, Kramer JJ. Spontaneous use of mnemonics by elderly individuals. Educ Gerontol 1983;9:57-71.
13Lachman ME, Steinberg ES, Trotter SD. Effects of control beliefs and attributions on memory self-assessments and performance. Psychol Aging 1987;2:266-71.
14Berry JM, West RL, Dennehey DM. Reliability and validity of the memory self-efficacy questionnaire. Dev Psychol 1989;25:701-13.
15Devolder PA, Brigham MC, Pressley M. Memory performance awareness in younger and older adults. Psychol Aging 1990;5:291-303.
16Moulin CJA, Perfect TJ, Jones RW. Global predictions of memory in Alzheimer's disease: Evidence for preserved metamemory monitoring. Aging, Neuropsychol Cogn 2000;7:230-44.
17Tiwari SC, Kumar A, Kumar A. Development & standardization of a scale to measure socio-economic status in urban & rural communities in India. Indian J Med Res 2005;122:309-14.
18Elias PK, Elias MF, D'Agostino RB, Silbershatz H, Wolf PA. Alcohol consumption and cognitive performance in the Framingham Heart Study. Am J Epidemiol 1999;150:580-9.
19Selvin E, Erlinger TP. Prevalence of and risk factors for peripheral arterial disease in the United States: Results from the National Health and Nutrition Examination Survey, 1999-2000. Circulation 2004;110:738-43.
20Selvin E, Erlinger TP. Prevalence of and risk factors for peripheral arterial disease in the United States: Results from the National Health and Nutrition Examination Survey, 1999–2000. Circulation 2004;110:738-43.
21D'cruz SM, Rajaratnam N. The effect of aging on cognitive function in a South Indian Population. Int J Sci Res Publ 2013;3:1-5.
22Murphy MD, Schmitt FA, Caruso MJ, Sanders RE. Metamemory in older adults: The role of monitoring in serial recall. Psychol Aging 1987;2:331-9.
23Schunn CD, Reder LM, Nhouyvanisvong A, Richards DR, Stroffolino PJ. To calculate or not to calculate: A source activation confusion model of problem familiarity's role in strategy selection. J Exp Psychol Learn Mem Cogn 1997;23:3-29.
24Dunlosky J, Nelson TO. Importance of the kind of cue for judgments of learning (JOL) and the delayed-JOL effect. Mem Cognit 1992;20:374-80.
25Lovelace EA, Marsh GR. Prediction and evaluation of memory performance by young and old adults. J Gerontol 1985;40:192-7.
26Salthouse TA. Are individual differences in rates of aging greater at older ages? Neurobiol Aging 2012;33:2373-81.
27Hertzog C, Dunlosky J. Metacognition in later adulthood: Spared monitoring can benefit older adults' self-regulation. Curr Dir Psychol Sci 2011;20:167-73.