Experimental Effects of Acute Exercise on Episodic Memory Function: Considerations for the Timing of Exercise

Abstract

Our previous work employing a between-subject randomized controlled trial design suggests that exercising prior to memory encoding is more advantageous in enhancing retrospective episodic memory function when compared to exercise occurring during or after memory encoding. The present experiment evaluates this potential temporal effect of acute exercise on memory function while employing a within-subject, counterbalanced design. In a counterbalanced order (via Latin squares), 24 participants completed four visits including (1) exercising (moderate-intensity walking) prior to memory encoding, (2) exercising during memory encoding, (3) exercising after memory encoding, and (4) a control visit (no exercise). Retrospective memory function (short term and long term; 24-hour follow-up) was assessed from a multitrial word list. Prospective memory was assessed from a time-based task. Compared to all other visits, short-term memory was greater in the visit that involved exercising prior to memory encoding (F = 3.76; P = .01; η²= .79). Similar results occurred for long-term memory, with no significant effects for prospective memory performance. We provide robust evidence demonstrating that acute moderate-intensity exercise prior to memory encoding is optimal in enhancing short-term and long-term memory function when compared to no exercise as well as exercising during and after memory encoding.

Keywords

Acquisition cognition consolidation encoding episodic physical activity

Introduction

Extensive research suggests that acute exercise is capable of enhancing cognitive performance outcomes (Chang, Labban, Gapin, & Etnier, 2012; Cox et al., 2016; Lambourne & Tomporowski, 2010; Roig, Nordbrandt, Geertsen, & Nielsen, 2013). Exercise is posited to modulate various correlates of optimal cognitive functioning among young to middle-aged adults as well, despite mixed evidence. Our recent work demonstrates that acute exercise can help to facilitate episodic retrospective memory (Crush & Loprinzi, 2017; Frith, Sng, & Loprinzi, 2017; Loprinzi, Frith, Edwards, Sng, & Ashpole, 2017; Loprinzi & Kane, 2015; Sng, Frith, & Loprinzi, 2018), with mechanisms of this effect discussed elsewhere (Loprinzi, Edwards, & Frith, 2017). Of interest to this study (written as a brief report) is the potential temporal effect of acute exercise on memory function. That is, the extent to which acute exercise may have a differential effect on episodic memory function depending on when the bout of exercise occurs relative to the learning task (Labban & Etnier, 2011). We evaluated this in two of our previous experiments (Frith et al., 2017; Sng et al., 2017) by examining the effects of acute exercise prior to memory encoding, during memory encoding, and during memory consolidation. One study (Sng et al., 2017) employed a moderate-intensity bout of acute exercise, while the other experiment (Frith et al., 2017) employed a high-intensity bout of acute exercise. Both of these previous experiments (Frith et al., 2017; Sng et al., 2017) demonstrated that exercise prior to memory encoding was more advantageous in facilitating learning and long-term memory when compared to exercising during the other phases of memory formation.

This experiment extends our previous two experiments (Frith et al., 2017; Sng et al., 2017), which were between-subject, randomized controlled trials. This study examined the same research question but employed a within-subject, counterbalanced design. A major advantage of a within-subject (vs. between-subject) design is that it reduces error variance associated with individual differences (Thompson & Campbell, 2004). Additionally, if our previous hypothesis is again supported utilizing the present within-subject experimental design, then these findings will provide robust evidence of a causal effect of acute exercise on episodic memory (while considering the temporal effects of exercise).

Methods

Study design and participants

This study was approved by the University’s institutional review board, and all participants provided written consent prior to participation. The methodology of this study is nearly identical to our previous experiments, but here we employ a within-subject, counterbalanced (via Latin squares) study design. In brief, 24 participants (college students; ages 18–35 years) were recruited via a non-probability sampling approach (classroom announcements and word-of-mouth). Participants were ineligible for this study if they self-reported being a current smoker, had a concussion in the past 30 days, were pregnant, currently taking medication to regulate mood, took marijuana or other illicit substances in the past two days, or had a diagnosis of attention deficit disorder/attention deficit hyperactivity disorder or a learning disability. Further, for any of the visits, if they exercised 5 hours prior to the visit or consumed caffeine 3 hours prior, the visit was rescheduled.

Participants completed four visits in our Exercise Psychology Laboratory. The four visits included the following: (1) exercising prior to memory encoding, (2) exercising during memory encoding, (3) exercising after memory encoding, and (4) a control visit (no exercise). Each visit took approximately 1 hour. After each visit, participants contacted the researcher via phone 24 hours later for a follow-up assessment. This 24-hour follow-up assessment served as an assessment of prospective memory (whether they remembered to call or not) and long-term retrospective episodic memory function. This laboratory visit plus 24-hour follow-up cycle was performed four separate times. Subsequent visits occurred at approximately the same time of day as the initial visit (±2 hours) for all participants. The time period between each cycle was at least 24 hours.

Measures

Survey

At the beginning of each visit, to assess mood status, participants completed the Positive and Negative Affect Schedule (Watson, Clark, & Tellegen, 1988). For this mood survey, participants rated 20 items (e.g., excited, upset, irritable, attentive) on a Likert scale (1, very slightly or not at all; to 5, extremely), with half of the items constituting a “positive” mood state, with the other half being a “negative” mood state. As a measure of habitual physical activity behavior, and reported as time spent per week in moderate-to-vigorous physical activity, participants also completed a survey (Physical Activity Vital Signs Questionnaire) at the beginning of the first visit (Ball, Joy, Gren, & Shaw, 2016). Also, during the first visit, waist circumference, and height and weight (to calculate body mass index) were measured to provide anthropometric characteristics of the sample.

Prospective memory

After each laboratory visit, the participant and researcher identified an agreeable time for the 24-hour follow-up assessment. If the participant called the researcher at the agreed upon time (±5 minutes), they were given a success score (“1”), whereas those who failed to call the researcher at this time were given a failure score (“0”) (Frith et al., 2017). For those who failed this task, the researcher then called the participant for the completion of the phone-based 24-hour follow-up assessment. Notably, all 24-hour assessments were completed.

Retrospective memory

Short-term and long-term memory (retrospective memory) were assessed using the standardized Rey Auditory Verbal Learning Test (RAVLT) (Rey, 1941). Participants were asked to listen to and immediately recall a recording of a list of 15 words (List A) five times in a row (Trials 1–5). Each word list was recorded at a rate of approximately one word per second. Participants were then asked to listen to and immediately recall a list of 15 new words (List B). Immediately following the recall of List B, participants were required to recall the words from List A (Trial 6). Four different word lists of equal difficulty were used for the individual visits (Crawford, Stewart, & Moore, 1989; Geffen, Butterworth, Forrester, & Geffen, 1994; Lezak, 1983; Majdan, Sziklas, & Jones-Gotman, 1996).

Approximately 24 hours later, researchers conducted a follow-up phone call with participants to assess 24-hour attribution and recognition. Participants were read a list of 50 words, with some words from List A and List B, as well as 20 new, distractor words that were not from either list used in the RAVLT assessment. Participants were asked to verbally recall whether they remember the words from List A, List B, or neither list. A recognition point was given if they correctly identified having previously heard a word, and an attribution point was given if they correctly identified a word as being from List A, List B, or neither list.

Exercise

The researcher instructed participants to walk on a treadmill (Woodway treadmill) for 15 minutes and select an appropriate pace by saying “Please select a pace similar to one you would choose if you were late to class. Thus, it will not be a leisurely walk. Nor will it be a run.” The self-selected pace was maintained during the exercise bout (i.e., the speed did not vary). Both heart rate and rating of perceived exertion (range, 6–20) were recorded during the midpoint (7.5 minutes) of their exercise and at the end of their exercise (14:55). Participants walked at the same speed for all visits (excluding the control condition).

Exercise and memory procedures

Depending on the visit, participants either (1) exercised for 15 minutes, followed by a 5-minute resting (sitting) period, and then completed the RAVLT (exercise prior to memory encoding); (2) exercised for 15 minutes while completing the RAVLT (starting at minute 5) during the bout of exercise (exercise during memory encoding); (3) completed the RAVLT and then exercised for 15 minutes afterward (exercise after memory encoding); or (4) completed a control visit, involving resting for 5 minutes and then completing the RAVLT (no exercise). For all visits, exactly 20 minutes after completing the RAVLT, participants completed an uncued, free-recall of List A (20-minute recall), which was performed without hearing the recorded RAVLT list again. During this delay period, for distraction purposes, participants watched an online episode of the TV show, “The Office.” Notably, a different episode was viewed for the subsequent visits.

Statistical analysis

Statistical analyses were computed in SPSS (Version 24). For attribution, recognition, Trial 6, and the 20-minute delay assessments, a 1 (trial) × 4 (visits) repeated measures analysis of variance (ANOVA) was computed. For the prospective memory assessment, a 1 (trial) × 4 (visits) Cochran’s Q was computed. For Trials 1–5, a 4 (visit) × 5 (trials) repeated measures ANOVA was computed. Partial eta-squared (η²) effect size estimates were calculated for each analysis. Statistical significance was established as the arbitrary alpha level of .05.

Results

Characteristics of the sample are shown in Table 1. Participants, on average, were 20.9 years, with a greater proportion being female (66.7%) and non-Hispanic white (79.2%). The mean treadmill speed for all exercise trials was 3.4 mph. Participant mood state was similar across the four counterbalanced visits.

Table 1.

Characteristics of the sample (N = 24).

Variable	Point estimate	SD
Age, mean years	20.9	1.9
Gender, % female	66.7
Race, % non-Hispanic white	79.2
Waist circumference, mean inches	35.6	4.3
MVPA, mean min/week	118.3	96.3
Treadmill speed (same for all exercise visits), mean MPH	3.4	0.2
Resting heart rate, mean BPM
Exercise prior to memory encoding visit	87.6	14.5
Exercise during memory encoding visit	85.0	13.5
Exercise after memory encoding visit	84.0	15.0
End of exercise heart rate, mean BPM
Exercise prior to memory encoding visit	127.5	16.9
Exercise during memory encoding visit	124.3	17.7
Exercise after memory encoding visit	125.1	18.8
Heart rate immediately prior to RAVLT assessment
Exercise prior to memory encoding visit	101.6	19.3
Exercise during memory encoding visit	113.1	31.2
Exercise after memory encoding visit	90.1	18.1
Mood state, mean
Positive—prior to encoding	25.0	9.2
Positive—during to encoding	26.0	10.0
Positive—after encoding	27.2	9.1
Positive—control	28.2	9.1
Negative—prior to encoding	14.0	8.0
Negative—during to encoding	12.2	2.4
Negative—after encoding	14.1	7.2
Negative—control	14.4	6.9

Note: Regarding the mood state assessment (PANAS), participants rated 20 items (e.g., excited, upset, irritable, attentive) on a Likert scale (1, very slightly or not at all; to 5, extremely), with half of the items constituting a “positive” mood state, with the other half being a “negative” mood state. For both positive and negative mood, the range is 10–50, with a higher score indicative of a greater positive/negative mood. This mood assessment occurred at the start of each of the four visits (i.e., exercise prior to memory encoding, exercise during memory encoding, exercise after memory encoding, and the control visit). BPM: beats per minute; MPH, miles per hour; MVPA, moderate-to-vigorous physical activity; PANAS: Positive and Negative Affect Schedule; RAVLT: Rey Auditory Verbal Learning Test.

Table 2 displays the memory performances across the visits. For the first five trials (assessment of learning), a visit × trial interaction effect was observed (F = 3.76; P = .01; η²= .79), in that learning improved across the trials, with the greatest effect occurring during the visit that involved exercising prior to memory encoding (also see Figure 1). Similarly, memory for Trial 6 (assessment of memory after a retroactive interference; List B) was greatest among the visit involving exercise prior to memory encoding (F = 3.89; P = .02; η²= .35). A similar trend occurred for the 20-minute memory delay (F = 2.50; P = .08; η²= .26) and long-term memory attribution (F = 2.89; P = .06; η²= .29). No significant effect occurred for prospective memory (P > .05).

Table 2.

Memory performances (mean number of words; SD) across the experimental visits (N = 24).

Memory	Prior to encoding	During encoding	After encoding	Control	Test statistic
Trial 1	7.46 (2.4)	6.0 (1.7)	5.96 (1.9)	6.13 (1.5)	F(visit) = 4.12, P = .01, η²= .37 F(trial) = 78.03, P < .01, η²= .94 F(visit × trial) = 3.76, P = .01, η²= .79
Trial 2	9.33 (2.4)	8.67 (1.9)	8.75 (1.8)	9.5 (1.6)
Trial 3	10.96 (2.0)	10.83 (2.5)	10.25 (2.2)	11.1 (1.8)
Trial 4	12.17 (1.9)	12.21 (2.4)	10.75 (3.2)	11.75 (1.8)
Trial 5	13.17 (1.9)	12.71 (2.1)	12.29 (2.1)	12.3 (2.2)
Trial 6	12.21 (2.6)	11.75 (2.2)	10.71 (3.0)	10.9 (2.7)	F = 3.89; P = .02; η²= .35
20-minute delay	10.50 (3.9)	11.20 (2.9)	10.04 (3.1)	10.04 (3.3)	F = 2.50; P = .08; η²= .26
PM, % success	37.5	29.2	29.2	45.8	Cochran’s Q = 2.53; P = .46
Attribution	19.67 (5.1)	19.08 (5.0)	17.83 (5.6)	17.96 (6.8)	F = 2.89; P = .06; η²= .29
Recognition	22.52 (3.9)	22.48 (4.2)	21.57 (4.6)	22.26 (4.8)	F = 0.50; P = .68; η²= .07

Note: For attribution, recognition, Trial 6, and 20-minute delay assessments, a 1 (trial) × 4 (visits) repeated measures ANOVA was computed. For PM, a 1 (trial) × 4 (visits) Cochran’s Q was computed. For Trials 1–5, a 4 (visit) × 5 (trials) repeated measures ANOVA was computed. ANOVA: analysis of variance; PM: prospective memory; SD: standard deviation.

Figure 1.

Number of words recalled across the six trials.

Discussion

Previous work suggests that exercise may benefit cognitive performance via a variety of mechanisms, including, but not limited to, increased cerebral blood flow (Guiney, Lucas, Cotter, & Machado, 2015), neurogenesis, synaptic plasticity, cell proliferation, acute increases in peripheral brain-derived neurotrophic factor (Tsai et al., 2014), and associated neural efficiency, which may be partially dependent on cardiorespiratory fitness (Tsai, Pan, Chen, Wang, & Chou, 2016). Recent studies have demonstrated that acute exercise can enhance both short- and long-term memory function (Crush & Loprinzi, 2017; Frith et al., 2017; Loprinzi, Frith, et al., 2017; Loprinzi & Kane, 2015; Sng et al., 2017), with emerging research demonstrating that the timing of exercise may moderate this effect (Frith et al., 2017; Sng et al., 2017). That is, our recent work (Frith et al., 2017; Sng et al., 2017), employing a between-subject randomized controlled trial design, demonstrated that exercise (both moderate- and high-intensity exercise) prior to memory encoding was most advantageous in improving memory function compared to no exercise, as well as exercise during and after memory encoding. The purpose of this study was to reassess this potential temporal effect of acute exercise on episodic memory function while employing a within-subject, counterbalanced design. The findings from this within-subject design align with our previous between-subject design results. That is, in this experiment, the visit involving exercise prior to memory encoding was most advantageous in short-term memory, learning, and long-term memory. Taken together, these experiments provide robust evidence that, within this young adult population, completing a relatively short (15 minutes) bout of exercise prior to memory encoding may be facilitative in improving short- and long-term episodic memory. As discussed elsewhere (Loprinzi, Edwards, et al., 2017), this may be a result of exercise-related alterations in neuronal excitability, which may help to facilitate long-term potentiation. Future work may wish to reexamine this temporal effect among other populations, such as older adults at the risk of dementia or mild cognitive impairment, and consider other modalities of exercise, such as cycling, or non-ambulatory-based activities. Examination of the potential for an extant dose–response relationship to emerge across diverse populations and exercise modalities may offer further insight into specific exercise prescription for a variety of individuals, particularly those with cognitive and/or mobility concerns. Further, and consistent with emerging work (Crush & Loprinzi, 2017), it would also extend the literature if future work evaluates other temporal periods prior to exercise, such as exercising 30 minutes, 60 minutes, and 120 minutes prior to the memory stimulus. Such an exercise temporal effect is plausible given that long-term potentiation may last for several hours (Bliss & Lomo, 1973). It would also be worthwhile for future research to continue investigating whether these exercise-enhancing effects (of various intensities and modalities) on word-list-related memory tasks extend to other types of memory tasks (e.g., paragraph retention tasks) (Labban & Etnier, 2011).

Footnotes

Article Notes

References

Ball

T. J.

Joy

E. A.

Gren

L. H.

Shaw

J. M.

(2016) Concurrent validity of a self-reported physical activity “vital sign” questionnaire with adult primary care patients. Preventing Chronic Disease 13: E16doi:10.5888/pcd13.150228.

Bliss

T. V.

Lomo

(1973) Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path. The Journal of Physiology 232(2): 331–356.

Chang

Y. K.

Labban

J. D.

Gapin

J. I.

Etnier

J. L.

(2012) The effects of acute exercise on cognitive performance: A meta-analysis. Brain Research 1453: 87–101. doi:10.1016/j.brainres.2012.02.068.

Cox

E. P.

O’Dwyer

Cook

Vetter

Cheng

H. L.

Rooney

O’Connor

(2016) Relationship between physical activity and cognitive function in apparently healthy young to middle-aged adults: A systematic review. Journal of Science and Medicine in Sport 19(8): 616–628. doi:10.1016/j.jsams.2015.09.003.

Crawford

J. R.

Stewart

L. E.

Moore

J. W.

(1989) Demonstration of savings on the AVLT and development of a parallel form. Journal of Clinical and Experimental Neuropsychology 11(6): 975–981. doi:10.1080/01688638908400950.

Crush

E. A.

Loprinzi

P. D.

(2017) Dose-response effects of exercise duration and recovery on cognitive functioning. Perceptual and Motor Skills 124(6): 1164–1193. doi:10.1177/0031512517726920.

Frith

Sng

Loprinzi

P. D.

(2017) Randomized controlled trial evaluating the temporal effects of high-intensity exercise on learning, short-term and long-term memory, and prospective memory. The European Journal of Neuroscience 46(10): 2557–2564. doi:10.1111/ejn.13719.

Geffen

G. M.

Butterworth

Forrester

G. M.

Geffen

L. B.

(1994) Auditory verbal learning test components as measures of the severity of closed-head injury. Brain Injury 8(5): 405–411.

Guiney

Lucas

S. J.

Cotter

J. D.

Machado

(2015) Evidence cerebral blood-flow regulation mediates exercise-cognition links in healthy young adults. Neuropsychology 29(1): 1–9. doi:10.1037/neu0000124.

10.

Labban

J. D.

Etnier

J. L.

(2011) Effects of acute exercise on long-term memory. Research Quarterly for Exercise and Sport 82(4): 712–721. doi:10.1080/02701367.2011.10599808.

11.

Lambourne

Tomporowski

(2010) The effect of exercise-induced arousal on cognitive task performance: A meta-regression analysis. Brain Research 1341: 12–24. doi:10.1016/j.brainres.2010.03.091.

12.

Lezak

M. D.

(1983) Neuropsychological assessment, New York, NY: Oxford University Press.

13.

Loprinzi

P. D.

Edwards

M. K.

Frith

(2017) Potential avenues for exercise to activate episodic memory-related pathways: A narrative review. The European Journal of Neuroscience 46(5): 2067–2077. doi:10.1111/ejn.13644.

14.

Loprinzi

P. D.

Frith

Edwards

M. K.

Sng

Ashpole

(2018) The effects of exercise on memory function among young- to middle-age adults: Systematic review and recommendations for future research. American Journal of Health Promotion 32(3): 691–704.

15.

Loprinzi

P. D.

Kane

C. J.

(2015) Exercise and cognitive function: A randomized controlled trial examining acute exercise and free-living physical activity and sedentary effects. Mayo Clinic Proceedings 90(4): 450–460. doi:10.1016/j.mayocp.2014.12.023.

16.

Majdan

Sziklas

Jones-Gotman

(1996) Performance of healthy subjects and patients with resection from the anterior temporal lobe on matched tests of verbal and visuoperceptual learning. Journal of Clinical and Experimental Neuropsychology 18(3): 416–430. doi:10.1080/01688639608408998.

17.

Rey

(1941) The psychological examination in cases of traumatic encepholopathy. Archives de Psychologie 28: 215–285.

18.

Roig

Nordbrandt

Geertsen

S. S.

Nielsen

J. B.

(2013) The effects of cardiovascular exercise on human memory: A review with meta-analysis. Neuroscience and Biobehavioral Reviews 37(8): 1645–1666. doi:10.1016/j.neubiorev.2013.06.012.

19.

Sng

Frith

Loprinzi

P. D.

(2018) Temporal effects of acute walking exercise on learning and memory. American Journal of Health Promotion. 32(7), 1518–1525. doi:10.1177/0890117117749476.

20.

Thompson

V. A.

Campbell

J. I.

(2004) A power struggle: Between- vs. within-subjects designs in deductive reasoning research. Psychologia 47: 277–296.

21.

Tsai

C. L.

Chen

F. C.

Pan

C. Y.

Wang

C. H.

Huang

T. H.

Chen

T. C.

(2014) Impact of acute aerobic exercise and cardiorespiratory fitness on visuospatial attention performance and serum BDNF levels. Psychoneuroendocrinology 41: 121–131. doi:10.1016/j.psyneuen.2013.12.014.

22.

Tsai

C. L.

Pan

C. Y.

Chen

F. C.

Wang

C. H.

Chou

F. Y.

(2016) Effects of acute aerobic exercise on a task-switching protocol and brain-derived neurotrophic factor concentrations in young adults with different levels of cardiorespiratory fitness. Experimental Physiology 101(7): 836–850. doi:10.1113/EP085682.

23.

Watson

Clark

L. A.

Tellegen

(1988) Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology 54(6): 1063–1070.