Behavioral Neuroscience 2013. Vol. 127, No. 6. 932-935
© 2013 American Psychological Association 0735-7044/13/$12.00 DOI: 10.1037/a0034930
Differential Facilitative Effects of Glucose Administration on Stroop Task Conditions Karen R. Brandt, E, Leigh Gibson, and James M. Rackie University of Roehampton Previous research has demonstrated that glucose administration improves memory performance. These glucose facilitation effects have been most reliably demonstrated in medial temporal lobe tasks with the greatest effects found for cognitively demanding tasks. The aim of the proposed research was to first explore whether such effects might be demonstrated in a frontal lobe task. A second aim was to investigate whether any beneficial effects of glucose may arise more prominently under tasks of increasing cognitive demand. To achieve these aims, the Stroop Task was administered to participants and effects of a drink of glucose (25 g) were compared with an aspartame-sweetened control drink on performance in young adults. Results demonstrated that glucose ingestion significantly reduced RTs in the congruent and incongruent conditions. No effect on error rates was observed. Of most importance was the finding that this glucose facilitative effect was significantly greatest in the most cognitively demanding task, that is, the incongruent condition. The present results support the contention that the glucose facilitation effect is most robust under conditions of enhanced task difficulty and demonstrate that such benefits extend to frontal lobe function. Keywords: attention, frontal lobe function, glucose facilitation effect, reaction times, Stroop task
Glucose ingestion has produced beneficial effects on a range of medial temporal lobe tasks (Smith, Riby, van Eekelen & Foster, 2011). There have, however, been some discrepant findings suggesting that glucose does not always have facilitative effects in such tasks (Brandt, Siinram-Lea, Jenkinson & Jones, 2010). There is some suggestion that this discrepancy may reflect the operation of different task demands. In other words, for glucose ingestion to enhance task performance, the task needs to be sufficiently difficult to deplete blood glucose levels in task-relevant brain areas, which therefore benefit from glucose ingestion, consequently resulting in better performance (Scholey, Laing, & Kennedy, 2006). Support for this premise comes from research showing glucose facilitation effects in highly demanding tasks (Riby, McLaughlin, & Riby, 2008; Sünram-Lea, Foster, Durlach & Perez, 2002). Most glucose facilitation effects have been observed in medial temporal lobe tasks (Smith et al., 2011); however, there is some evidence for glucose benefits for frontal lobe function. Research has demonstrated that glucose ingestion enhanced frontal event related potential (ERP) effects (Riby et al., 2008) as well as Stroop task performance in older adults (Gagnon, Greenwood & Bherer, 2010). In a study of younger adults, there was no overall glucose facilitation of Stroop task performance, although a subanalysis revealed faster performance, specifically in the incongruent condition only in participants with initially rising blood glucose levels (Benton, Owens & Parker, 1994). The absence of a generic glucose
facilitation effect in that study may be attributable to the lack of a fasting requirement and use of a high glucose dose (75 g) which is triple the optimal glucose dose in young adults (Messier, 2004), and it has been argued that high doses may even impair performance (Gibson, 2007). The aim of the present research was therefore to investigate whether glucose could improve Stroop task performance in fasting young adults, using 25 g of glucose. In particular, the present research sought to demonstrate greater glucose facilitation effects in the incongruent compared with the control or the congruent conditions of the Stroop Task.
Method Sixty undergraduate students (46 female, 14 male; mean age = 19.7 years) from the University of Roehampton participated for course credit. The double-blind, placebo-controlled study received ethical approval, and all participants gave informed consent. Participants were not diabetic and had English as their first language. The experiment had a mixed factorial design: 2 (drink; aspartame vs. glucose) X 3 (condition: control vs. congruent vs. incongruent) with repeated measures on the second factor. Participants received either 25 g glucose or 5 tablets of aspartame dissolved in 250 ml of water with two squirts of lemon. The Stroop task (Stroop, 1935) consisted of three conditions: a word task (control), a color task (congruent), and a color-word task (incongruent) all presented on an A4 sheet of paper. The word task consisted of 30 color words printed in black ink. The color task consisted of 30 colored squares, and the color-word task contained 30 color name words printed in a color different from the color name spelled. The words in the word and color-word task were in bold capital letters in 28 point Calibri font.
Karen R. Brandt, E. Leigh Gibson, and James M. Raekie, Department of Psychology, Whitelands College, University of Roehampton, London, UK. Correspondence concerning this article should be addressed to Karen R. Brandt, Department of Psychology, Whitelands College, University of Roehampton, Holybourne Avenue, London, SW15 4JD UK. E-mail: [email protected] 932
GLUCOSE AND STROOP
Participants arrived at the laboratory for a 45-min session between 0900h and 1200h after fasting overnight, and were randomly assigned to drink condition. Participants gave their age and were measured for weight and height before finger-prick blood glucose measurement (BGLl). All participants then received either a glucose- or aspartame-containing drink, rated the sweetness of the drink, and indicated how many sweet drinks per day they consumed. After a 15-min delay, the second blood glucose measurement (BGL2) was taken before the start of the Stroop Task. All participants were first given the control task which consisted of reading the list of words. Next, half the participants in each condition were administered the congruent task followed by the incongruent task, and vice versa for the other half. The congruent task consisted of naming the color of the printed squares, and the incongruent task involved naming the color of the printed word (rather than the color the word named). In all three conditions, participants were informed to complete the task as quickly and accurately as possible. Participants' total RTs to completing each of the three tasks and number of errors made were recorded. After the completion of the Stroop task, the third blood glucose measurement (BGL3) was taken, and participants were thanked, debriefed, and dismissed. Participants were excluded if BMI was below 18.5 or above 30 (mean BMI = 22.7), BGLl > 5.5, and they consumed at least 4 sweet drinks per day. Thus, 39 participants' data were included. Group differences in sweetness ratings and sweet drinks consumed were tested by independent t test, as were reaction time (RT) difference scores when examining drink X condition effects. Blood glucose levels, RTs and error rates were submitted separately to 2 (drink: aspartame vs. glucose) X 3 (time or repeated conditions) mixed repeated measures ANOVA. Significance levels are given as 2-tailed unless specified. In addition, because Benton et al., (1994) found only a significant infiuence of increasing versus decreasing pretask blood glucose levels on their Stroop task, and not of drink type, we conducted the same ANOVAs of Stroop performance on a reallocated sample to examine the importance of the change in blood glucose. However, unlike in Benton et al.'s unfasted participants (tested between 0900h and 1200h), few of our participants showed an absolute decrease in blood glucose 15 minutes postdrink and before the Stroop test. Therefore, we grouped participants into those whose pretest change in glucose from baseline was less than the overall mean and those whose blood glucose had increased above the mean. This resulted in the reallocation of 4 participants from the glucose drink condition into the "low change" group (n = 22), and 3 participants from the control drink condition into the "high change" group (n = 17).
933
Glycémie Response Blood glucose levels were greater overall in the glucose (M = 5.94) than the aspartame group (M = 4.87) [main effect of drink: F(l, 37) = 16.39, p < .001]. Pairwise comparisons (Bonferroni corrected) on the main effect of time, F(2, 74) = 54.78, p < .001, showed greater blood glucose levels at both BGL2 (M = 5.65) and BGL3 (M = 6.01) than BGL1(M = 4.54) as well as greater levels at BGL3 than BGL2. Finally a significant drink X time interaction was found, F(2, 74) = 17.90, p < .001, with further independent one-tailed t tests demonstrating no significant differences between groups at BGLl, i(37) = .24, p = .81, but significant differences at BGL2, i(37) = 4.99, p < .001, and BGL3, i(37) = 4.25, p < .001 (see Figure 1).
Reaction Times There was a significant main effect of condition on overall RTs, F(2, 74) = 279.91, p < .001. Pairwise comparisons (Bonferroni corrected) revealed that RTs in the control condition (M = 49.69) were significantly shorter than the congruent and incongruent conditions (respective Ms: 53.01 vs. 84.28) and that the congruent condition produced significantly shorter RTs than the incongruent condition. The main effect of drink just failed to reach significance, F(l, 37) = 3.08,p = .08; however, a significant interaction between condition and drink was found, F(2, 74) = 3.51, ;? < .05 (see Table 1). Independent one-tailed t tests revealed that RTs in the glucose group were significantly shorter than the aspartame group in the congruent condition, i(37) = 1.75, p < .05, as well as the incongruent condition, f(37) = 1.96, p < .05. To examine whether there was a differential facultative effect of glucose administration over aspartame across Stroop conditions, RT differences between the control and congruent conditions (congruent effect), as well as between the control and incongruent conditions (incongruent effect) were calculated for both glucose and aspartame groups. There were no significant differences between the glucose (M = 1.43) and the aspartame group (M = 5.21) for the congruent effect, i(37) = 1.86, p = .07, but a significant
8.0
7.0
" • ~
Glucose drink
- O-
Control drink
6.0
5.0
Results Sweetness Ratings/Drinks
o o
m
4.0
3.0
t15
No significant differences between the glucose (M = 39.44) and aspartame group (M = 44.14) emerged for participants' sweetness ratings, /(37) = .12, p = .47. Additionally, no differences between the glucose (M = 14.11) and aspartame groups (M = 20.19) were found in the number of sweet drinks consumed, f(37 = 1.07, p = .29.
t30
Time (mins) Figure 1. Capillary blood glucose levels by drink condition (glucose = solid circles; control = open circles) at baseline (tO), 15 minutes postdrink (tl5), and immediately after the Stroop task (t30). Data are expressed as mean ± SE.
934
BRANDT, GIBSON, AND RACKIE
Table 1 Total Reaction Times (S) and Error Rates as a Function of Condition and Task Stroop task
Reaction times Condition Glucose Aspartame Error rates Condition Glucose Aspartame Note.
Control
Congruent
Incongnient
48.82(1.95) 50.56(1.80)
50.25 (2.30) 55.77(2.13)
79.31 (3.70) 89.25 (3.42)
.33 (.23) .90 (.21)
3.27 (.90) 3.95 (.83)
.44 (.27) .76 (.25)
Data are means with standard errors in parentheses.
effect was found between the glucose (M = 30,49) and the aspartame group (M = 38.69) for the incongnient effect, r(37) = 2.14, p < .05. For the secondary analysis based on allocafion by glucose increasing or decreasing (above or below the mean) at 115, the only significant effect to emerge was for task condition, with RTs becoming longer with increasing task demand, as above, F(2, 74) = 265.23, p < .001. Thus, Stroop performance was not significantly dependent on relafive change in pretest blood glucose in these fasted participants.
Error Rates Analysis of error rates revealed a significant main effect of Stroop condition, F(2, 74) = 28.95, p < .001. Pairwise comparisons (Bonferroni corrected) revealed that error rates in the incongruent condifion {M = 3.61) were significantly greater than both the control and congruent condifions (respective Ms: .60 vs. .61). No other effects emerged. Finally, the only significant effect of error rates from ANOVA of relative change in pretest blood glucose was for the main effect of Stroop condifion, that is, more errors in the incongruent task, F(2, 74) = 28,20, p < ,001.
Discussion The aim of the present research was to invesfigate whether glucose facilitafion effects (i.e., improved cognifion following glucose administration) could be observed in young fasfing adults using age-appropriate glucose doses in the Stroop task. The results revealed that glucose administration significanfiy reduced RTs in the congruent and incongruent condifions with no effects on error rates observed. This replicates past research that has shown glucose to decrease RTs across a range of cognitive tasks (Donohoe & Benton, 1999; Green, Taylor, Elliman & Rhodes, 2001) as well as the Stroop task in older adults (Gagnon, Greenwood, & Bherer, 2010). Most importantly, the results demonstrated that the facilitafive effect of glucose was greatest in the most cognitively demanding task, that is, the incongruent condition. These results extend those found in older adults (Gagnon, Greenwood, & Bherer, 2010) where it was demonstrated that glucose facilitation effects occurred for the most demanding task conditions involving inhibition and switching, but not for the easier tasks of color-naming and
reading. In addifion, the present results demonstrated that, with the appropriate dose after an overnight fast, such effects can be extended to younger adults. Previous research using the Stroop task on younger adults that were not fasted, and where participants received triple the recommended glucose dose for opfimal task performance (Benton, Owens, & Parker, 1994), failed to find a generic glucose facilitation effect (despite finding improved word recall using the same drink treatment regime). In light of the present results, it is possible that the earlier negative finding was attributable to the use of a high glucose dose, thereby impairing performance in these participants on the more demanding task (Gibson, 2007). In our fasting participants, we also found no support for the argument that the change in blood glucose from baseline to itnmediately before the task was per se predictive of Stroop performance: rather, it was ingestion of glucose that produced the significant benefit. It should be noted that finger-tip blood glucose levels are not direct indicators of glucose uptake to task-specific areas of the brain and may be influenced by a variety of hormones, autonomie nervous system acfivity, and other physiological processes, which themselves may be responding to parficular nutritional and psychological states of the participants and could interact with performance (Gibson, 2007). This research provides further credence therefore to the view that different glucose doses are necessary across different populafion groups and test contexts to reliably observe glucose facilitafion effects (Messier, 2(X)4). In contrast to Benton, Owens, and Parker, (1994), we found no evidence that changes in pretest blood glucose levels were related to performance. Nevertheless, Gagnon, Greenwood, and Bherer (2011) recently reported that better glucoregulators (lower blood glucose area under the curves over 90 minutes) specifically made more errors during the most demanding component (switching) of a Stroop task, in agreement with previous findings (Awad, Gagnon, Desrochers, Tsiakas, and Messier, 2002; Lamport, Lawton, Mansfield & Dye, 2009; Messier, 2004; Messier, Awad-Shimoon, Gagnon, Desrochers and Tsiakas, 2011), We were unable to test this relationship here because blood glucose was only measured for 30 minutes after the drink, that is, before declining glucose levels. The finding that the greatest glucose facilitation effects arose under the most cognitively demanding conditions in the present research additionally supports the contention that glucose facilitation effects are most robust under conditions of sufficient task difficulty to cause depletion of circulating blood glucose levels in task-relevant brain areas (Scholey, Laing, & Kennedy, 2006; Siinram-Lea et al, 2002). Although this hypothesis is supported for hippocampal tasks by direct observations of changes in cerebral glucose in rats, it is not clear whether such sensitivity exists for frontal lobe tasks (McNay, Fries, & Gold, 2000). Moreover, delivery of glucose from blood to active neurones is normally exquisitely buffered by astrocytes, to allow supply to meet demand (Peters et al, 2004), The lack of glucose facilitafion effects with higher doses (Smith et al, 2011) is also hard to explain in terms of repletion of glucose during demanding tasks, although such doses may provoke substantial cortisol release that is associated with memory impairment (Gibson, 2007), Finally, the present results demonstrated that glucose facilitafion effects can be observed for nonmedial temporal lobe tasks. Most research demonstrating glucose enhancement effects have done so using episodic memory tasks mediated by different areas of the
GLUCOSE AND STROOP medial temporal lobes such as the hippocampus (Riby et al., 2008; Smith et al., 2011; Sünram-Lea, Dewhurst & Foster, 2008) or the parahippocampus (Stone, Thermenos, Tarbox, Poldrack & Seidman, 2005). However, there is some demonstration that the beneficial effects of glucose administration can extend beyond the medial temporal lobes. For example, research has shown a trend toward enhanced activation after glucose administration in the left dorsolateral prefrontal cortex in schizophrenics (Stone et al., 2005). In addition, Riby et al. (2008) found some evidence that the P3a ERP frontal lobe component was enhanced after glucose administration in normal young adults, although they contended that variability in their data and a small sample size meant further work in this area is required. Finally Gagnon, Greenwood, and Bherer (2010) showed enhanced Stroop performance in 60- to 80-year-old adults, albeit with twice the dose of glucose used in the present research. To our knowledge, therefore, our results provide the first support for a reliable glucose facilitation effect in frontal lobe function in young adults and at the same low dose that appears to be most effective in enhancing medial temporal lobe function. One clear task for future research would be to investigate whether other frontal lobe tasks benefit from glucose ingestion and whether such benefits are of similar magnitude to those enjoyed by medial temporal lobes tasks.
References Awad, N., Gagnon, M., Desrochers, A., Tsiakas, M., & Messier, C. (2002). Impact of peripheral glucoregulation on memory. Behavioral Neuroscience, 116, 691-702. doi:10.1037/0735-7044.116.4.691 Benton, D., Owens, S., & Parker, P. Y. (1994). Blood glucose influences memory and attention in young adults. Neuropsychologia, 32, 595-607. doi: 10.1016/0028-3932(94)90147-3 Brandt, K. R., SUnram-Lea, S., Jenkinson, P. M., & Jones, E. (2010). Glucose and emotional memory: Moderating effects of dose and task difficulty. Behavioural Brain Research, 211, 83-88. doi:10.1016/j.bbr .2010.03.016 Donohoe, R. T., & Benton, D. (1999). Cognitive functioning is susceptible to the level of blood glucose. Psychopharmacology, 145, 378-385. doi:10.1007/s002130051071 Gagnon, C , Greenwood, C. E., & Bherer, L. (2010). The acute effects of glucose ingestion on attentional control in fasting healthy older adults. Psychopharmacoiogy, 211, 337-346. doi:10.1007/s00213-010-1905-9 Gagnon, C , Greenwood, C. E., & Bherer, L. (2011). Glucose regulation is associated with attentional control performances in nondiabetic older adults. Joumai of Clinical and Experimental Neuropsychology, 33, 972-981. doi:10.1080/13803395.2011.589372 Gibson, E. L. (2007). Carbohydrates and mental function: Feeding or impeding the brain? Nutrition Bulletin, 32, 71-83. doi:10.1 lll/j.14673010.2007.00606.x Green, M. W., Taylor, M. A., Elliman, N. A., & Rhodes, O. (2001). Placebo expectancy effects in the relationship between glucose and cognition. British Journal of Nutrition, 86, 173-179. doi:10.1079/ BJN2001398
935
Lamport, D. J., Lawton, C. L., Mansfield, M. W., & Dye, L. (2009). Impairments in glucose tolerance can have a negative impact on cognitive function: A systematic research review. Neuroscience and Biobeiiavioral Reviews, 33, 394-413. doi:10.1016/j.neubiorev.2008.10.008 McNay, E. C, Fries, T. M., & Gold, P. E. (2000). Decreases in rat extracellular hippocampal glucose concentration associated with cognitive demand during a spatial task. Proceedings of the National Academy of Sciences of the United States of America, 97, 2881-2885. doi: 10.1073/pnas.050583697 Messier, C. (2004). Glucose and memory, a review. European Journai of Pharmacology, 490, 33-57. doi:10.1016/j.ejphar.2004.02.043 Messier, C , Awad-Shimoon, N., Gagnon, M., Desrochers, A., & Tsiakas, M. (2011). Glucose regulation is associated with cognitive performance in young nondiabetic adults. Behavioural Brain Research, 222, 81-88. doi:10.1016/j.bbr.2011.03.023 Peters, A., Schweiger, U., Pellerin, L., Hubold, C , Oltmanns, K. M., Conrad, M., . . . Fehm, H. L. (2004). The selfish brain: Competition for energy resources. Neuroscience and Biobehaviorai Reviews, 28, 143180. doi:10.1016/j.neubiorev.2004.03.002 Riby, L., McLaughlin, J., & Riby, D. (2008). Lifestyle, glucose regulation and the cognitive effects of glucose load in middle-aged adults. The British Journal of Nutrition, 100, 1128-1134. Riby, L. M., Sünram-Lea, S. L, Graham, C , Foster, J. K., Cooper, T., Moodie, C , & Gunn, V. P. (2008). P3b versus P3a: An event-related potential investigation of the glucose facilitation effect. Joumal of Psychopharmacoiogy, 22, 486-492. doi:10.1177/0269881107081561 Scholey, A. B., Laing, S., & Kennedy, D. O. (2006). Blood glucose changes and memory: Effects of manipulating emotionality and mental effort. Biologicai Psychoiogy, 71, 12-19. doi:10.1016/j.biopsycho.2005 .02.003 Smith, M. A., Riby, L. M., van Eekelen, J. A. M., & Foster, J. K. (2011). Glucose enhancement of human memory: A comprehensive research review of the glucose memory facilitation effect. Neuroscience and Biobehaviorai Reviews, 35, 770-783. doi:10.1016/j.neubiorev.2010.09 .008 Stone, W. S., Thermenos, H. W., Tarbox, S. I., Poldrack, R. A., & Seidman, L. J. (2005). Medial temporal and prefrontal lobe activation during verbal encoding following glucose ingestion in schizophrenia: A pilot fMRI study. Neurobiology of learning and Memory, 83, 54-64. doi:10.1016/j.nlm.2004.07.009 Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18, 643-662. doi:10.1037/ h0054651 Sünram-Lea, S. I.. Dewhurst, S. A., & Foster, J. K. (2008). The effect of gluco.se administration on the recollection and familiarity components of recognition memory. Biological Psychology, 77, 69-75. doi:10.1016/j .biopsycho.2007.09.006 Sünram-Lea, S. I., Foster, J. K., Durlach, P., & Perez, C. (2002). Investigation into the significance of task difficulty and divided allocation of resources on the glucose memory facilitation effect. Psychopharmacology, 160, 387-397. doi:10.1007/s00213-001-0987-9 Received July 22, 2013 Revision received October 3, 2013 Accepted October 5, 2013 •
Copyright of Behavioral Neuroscience is the property of American Psychological Association and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
© 2013 American Psychological Association 0735-7044/13/$12.00 DOI: 10.1037/a0034930
Differential Facilitative Effects of Glucose Administration on Stroop Task Conditions Karen R. Brandt, E, Leigh Gibson, and James M. Rackie University of Roehampton Previous research has demonstrated that glucose administration improves memory performance. These glucose facilitation effects have been most reliably demonstrated in medial temporal lobe tasks with the greatest effects found for cognitively demanding tasks. The aim of the proposed research was to first explore whether such effects might be demonstrated in a frontal lobe task. A second aim was to investigate whether any beneficial effects of glucose may arise more prominently under tasks of increasing cognitive demand. To achieve these aims, the Stroop Task was administered to participants and effects of a drink of glucose (25 g) were compared with an aspartame-sweetened control drink on performance in young adults. Results demonstrated that glucose ingestion significantly reduced RTs in the congruent and incongruent conditions. No effect on error rates was observed. Of most importance was the finding that this glucose facilitative effect was significantly greatest in the most cognitively demanding task, that is, the incongruent condition. The present results support the contention that the glucose facilitation effect is most robust under conditions of enhanced task difficulty and demonstrate that such benefits extend to frontal lobe function. Keywords: attention, frontal lobe function, glucose facilitation effect, reaction times, Stroop task
Glucose ingestion has produced beneficial effects on a range of medial temporal lobe tasks (Smith, Riby, van Eekelen & Foster, 2011). There have, however, been some discrepant findings suggesting that glucose does not always have facilitative effects in such tasks (Brandt, Siinram-Lea, Jenkinson & Jones, 2010). There is some suggestion that this discrepancy may reflect the operation of different task demands. In other words, for glucose ingestion to enhance task performance, the task needs to be sufficiently difficult to deplete blood glucose levels in task-relevant brain areas, which therefore benefit from glucose ingestion, consequently resulting in better performance (Scholey, Laing, & Kennedy, 2006). Support for this premise comes from research showing glucose facilitation effects in highly demanding tasks (Riby, McLaughlin, & Riby, 2008; Sünram-Lea, Foster, Durlach & Perez, 2002). Most glucose facilitation effects have been observed in medial temporal lobe tasks (Smith et al., 2011); however, there is some evidence for glucose benefits for frontal lobe function. Research has demonstrated that glucose ingestion enhanced frontal event related potential (ERP) effects (Riby et al., 2008) as well as Stroop task performance in older adults (Gagnon, Greenwood & Bherer, 2010). In a study of younger adults, there was no overall glucose facilitation of Stroop task performance, although a subanalysis revealed faster performance, specifically in the incongruent condition only in participants with initially rising blood glucose levels (Benton, Owens & Parker, 1994). The absence of a generic glucose
facilitation effect in that study may be attributable to the lack of a fasting requirement and use of a high glucose dose (75 g) which is triple the optimal glucose dose in young adults (Messier, 2004), and it has been argued that high doses may even impair performance (Gibson, 2007). The aim of the present research was therefore to investigate whether glucose could improve Stroop task performance in fasting young adults, using 25 g of glucose. In particular, the present research sought to demonstrate greater glucose facilitation effects in the incongruent compared with the control or the congruent conditions of the Stroop Task.
Method Sixty undergraduate students (46 female, 14 male; mean age = 19.7 years) from the University of Roehampton participated for course credit. The double-blind, placebo-controlled study received ethical approval, and all participants gave informed consent. Participants were not diabetic and had English as their first language. The experiment had a mixed factorial design: 2 (drink; aspartame vs. glucose) X 3 (condition: control vs. congruent vs. incongruent) with repeated measures on the second factor. Participants received either 25 g glucose or 5 tablets of aspartame dissolved in 250 ml of water with two squirts of lemon. The Stroop task (Stroop, 1935) consisted of three conditions: a word task (control), a color task (congruent), and a color-word task (incongruent) all presented on an A4 sheet of paper. The word task consisted of 30 color words printed in black ink. The color task consisted of 30 colored squares, and the color-word task contained 30 color name words printed in a color different from the color name spelled. The words in the word and color-word task were in bold capital letters in 28 point Calibri font.
Karen R. Brandt, E. Leigh Gibson, and James M. Raekie, Department of Psychology, Whitelands College, University of Roehampton, London, UK. Correspondence concerning this article should be addressed to Karen R. Brandt, Department of Psychology, Whitelands College, University of Roehampton, Holybourne Avenue, London, SW15 4JD UK. E-mail: [email protected] 932
GLUCOSE AND STROOP
Participants arrived at the laboratory for a 45-min session between 0900h and 1200h after fasting overnight, and were randomly assigned to drink condition. Participants gave their age and were measured for weight and height before finger-prick blood glucose measurement (BGLl). All participants then received either a glucose- or aspartame-containing drink, rated the sweetness of the drink, and indicated how many sweet drinks per day they consumed. After a 15-min delay, the second blood glucose measurement (BGL2) was taken before the start of the Stroop Task. All participants were first given the control task which consisted of reading the list of words. Next, half the participants in each condition were administered the congruent task followed by the incongruent task, and vice versa for the other half. The congruent task consisted of naming the color of the printed squares, and the incongruent task involved naming the color of the printed word (rather than the color the word named). In all three conditions, participants were informed to complete the task as quickly and accurately as possible. Participants' total RTs to completing each of the three tasks and number of errors made were recorded. After the completion of the Stroop task, the third blood glucose measurement (BGL3) was taken, and participants were thanked, debriefed, and dismissed. Participants were excluded if BMI was below 18.5 or above 30 (mean BMI = 22.7), BGLl > 5.5, and they consumed at least 4 sweet drinks per day. Thus, 39 participants' data were included. Group differences in sweetness ratings and sweet drinks consumed were tested by independent t test, as were reaction time (RT) difference scores when examining drink X condition effects. Blood glucose levels, RTs and error rates were submitted separately to 2 (drink: aspartame vs. glucose) X 3 (time or repeated conditions) mixed repeated measures ANOVA. Significance levels are given as 2-tailed unless specified. In addition, because Benton et al., (1994) found only a significant infiuence of increasing versus decreasing pretask blood glucose levels on their Stroop task, and not of drink type, we conducted the same ANOVAs of Stroop performance on a reallocated sample to examine the importance of the change in blood glucose. However, unlike in Benton et al.'s unfasted participants (tested between 0900h and 1200h), few of our participants showed an absolute decrease in blood glucose 15 minutes postdrink and before the Stroop test. Therefore, we grouped participants into those whose pretest change in glucose from baseline was less than the overall mean and those whose blood glucose had increased above the mean. This resulted in the reallocation of 4 participants from the glucose drink condition into the "low change" group (n = 22), and 3 participants from the control drink condition into the "high change" group (n = 17).
933
Glycémie Response Blood glucose levels were greater overall in the glucose (M = 5.94) than the aspartame group (M = 4.87) [main effect of drink: F(l, 37) = 16.39, p < .001]. Pairwise comparisons (Bonferroni corrected) on the main effect of time, F(2, 74) = 54.78, p < .001, showed greater blood glucose levels at both BGL2 (M = 5.65) and BGL3 (M = 6.01) than BGL1(M = 4.54) as well as greater levels at BGL3 than BGL2. Finally a significant drink X time interaction was found, F(2, 74) = 17.90, p < .001, with further independent one-tailed t tests demonstrating no significant differences between groups at BGLl, i(37) = .24, p = .81, but significant differences at BGL2, i(37) = 4.99, p < .001, and BGL3, i(37) = 4.25, p < .001 (see Figure 1).
Reaction Times There was a significant main effect of condition on overall RTs, F(2, 74) = 279.91, p < .001. Pairwise comparisons (Bonferroni corrected) revealed that RTs in the control condition (M = 49.69) were significantly shorter than the congruent and incongruent conditions (respective Ms: 53.01 vs. 84.28) and that the congruent condition produced significantly shorter RTs than the incongruent condition. The main effect of drink just failed to reach significance, F(l, 37) = 3.08,p = .08; however, a significant interaction between condition and drink was found, F(2, 74) = 3.51, ;? < .05 (see Table 1). Independent one-tailed t tests revealed that RTs in the glucose group were significantly shorter than the aspartame group in the congruent condition, i(37) = 1.75, p < .05, as well as the incongruent condition, f(37) = 1.96, p < .05. To examine whether there was a differential facultative effect of glucose administration over aspartame across Stroop conditions, RT differences between the control and congruent conditions (congruent effect), as well as between the control and incongruent conditions (incongruent effect) were calculated for both glucose and aspartame groups. There were no significant differences between the glucose (M = 1.43) and the aspartame group (M = 5.21) for the congruent effect, i(37) = 1.86, p = .07, but a significant
8.0
7.0
" • ~
Glucose drink
- O-
Control drink
6.0
5.0
Results Sweetness Ratings/Drinks
o o
m
4.0
3.0
t15
No significant differences between the glucose (M = 39.44) and aspartame group (M = 44.14) emerged for participants' sweetness ratings, /(37) = .12, p = .47. Additionally, no differences between the glucose (M = 14.11) and aspartame groups (M = 20.19) were found in the number of sweet drinks consumed, f(37 = 1.07, p = .29.
t30
Time (mins) Figure 1. Capillary blood glucose levels by drink condition (glucose = solid circles; control = open circles) at baseline (tO), 15 minutes postdrink (tl5), and immediately after the Stroop task (t30). Data are expressed as mean ± SE.
934
BRANDT, GIBSON, AND RACKIE
Table 1 Total Reaction Times (S) and Error Rates as a Function of Condition and Task Stroop task
Reaction times Condition Glucose Aspartame Error rates Condition Glucose Aspartame Note.
Control
Congruent
Incongnient
48.82(1.95) 50.56(1.80)
50.25 (2.30) 55.77(2.13)
79.31 (3.70) 89.25 (3.42)
.33 (.23) .90 (.21)
3.27 (.90) 3.95 (.83)
.44 (.27) .76 (.25)
Data are means with standard errors in parentheses.
effect was found between the glucose (M = 30,49) and the aspartame group (M = 38.69) for the incongnient effect, r(37) = 2.14, p < .05. For the secondary analysis based on allocafion by glucose increasing or decreasing (above or below the mean) at 115, the only significant effect to emerge was for task condition, with RTs becoming longer with increasing task demand, as above, F(2, 74) = 265.23, p < .001. Thus, Stroop performance was not significantly dependent on relafive change in pretest blood glucose in these fasted participants.
Error Rates Analysis of error rates revealed a significant main effect of Stroop condition, F(2, 74) = 28.95, p < .001. Pairwise comparisons (Bonferroni corrected) revealed that error rates in the incongruent condifion {M = 3.61) were significantly greater than both the control and congruent condifions (respective Ms: .60 vs. .61). No other effects emerged. Finally, the only significant effect of error rates from ANOVA of relative change in pretest blood glucose was for the main effect of Stroop condifion, that is, more errors in the incongruent task, F(2, 74) = 28,20, p < ,001.
Discussion The aim of the present research was to invesfigate whether glucose facilitafion effects (i.e., improved cognifion following glucose administration) could be observed in young fasfing adults using age-appropriate glucose doses in the Stroop task. The results revealed that glucose administration significanfiy reduced RTs in the congruent and incongruent condifions with no effects on error rates observed. This replicates past research that has shown glucose to decrease RTs across a range of cognitive tasks (Donohoe & Benton, 1999; Green, Taylor, Elliman & Rhodes, 2001) as well as the Stroop task in older adults (Gagnon, Greenwood, & Bherer, 2010). Most importantly, the results demonstrated that the facilitafive effect of glucose was greatest in the most cognitively demanding task, that is, the incongruent condition. These results extend those found in older adults (Gagnon, Greenwood, & Bherer, 2010) where it was demonstrated that glucose facilitation effects occurred for the most demanding task conditions involving inhibition and switching, but not for the easier tasks of color-naming and
reading. In addifion, the present results demonstrated that, with the appropriate dose after an overnight fast, such effects can be extended to younger adults. Previous research using the Stroop task on younger adults that were not fasted, and where participants received triple the recommended glucose dose for opfimal task performance (Benton, Owens, & Parker, 1994), failed to find a generic glucose facilitation effect (despite finding improved word recall using the same drink treatment regime). In light of the present results, it is possible that the earlier negative finding was attributable to the use of a high glucose dose, thereby impairing performance in these participants on the more demanding task (Gibson, 2007). In our fasting participants, we also found no support for the argument that the change in blood glucose from baseline to itnmediately before the task was per se predictive of Stroop performance: rather, it was ingestion of glucose that produced the significant benefit. It should be noted that finger-tip blood glucose levels are not direct indicators of glucose uptake to task-specific areas of the brain and may be influenced by a variety of hormones, autonomie nervous system acfivity, and other physiological processes, which themselves may be responding to parficular nutritional and psychological states of the participants and could interact with performance (Gibson, 2007). This research provides further credence therefore to the view that different glucose doses are necessary across different populafion groups and test contexts to reliably observe glucose facilitafion effects (Messier, 2(X)4). In contrast to Benton, Owens, and Parker, (1994), we found no evidence that changes in pretest blood glucose levels were related to performance. Nevertheless, Gagnon, Greenwood, and Bherer (2011) recently reported that better glucoregulators (lower blood glucose area under the curves over 90 minutes) specifically made more errors during the most demanding component (switching) of a Stroop task, in agreement with previous findings (Awad, Gagnon, Desrochers, Tsiakas, and Messier, 2002; Lamport, Lawton, Mansfield & Dye, 2009; Messier, 2004; Messier, Awad-Shimoon, Gagnon, Desrochers and Tsiakas, 2011), We were unable to test this relationship here because blood glucose was only measured for 30 minutes after the drink, that is, before declining glucose levels. The finding that the greatest glucose facilitation effects arose under the most cognitively demanding conditions in the present research additionally supports the contention that glucose facilitation effects are most robust under conditions of sufficient task difficulty to cause depletion of circulating blood glucose levels in task-relevant brain areas (Scholey, Laing, & Kennedy, 2006; Siinram-Lea et al, 2002). Although this hypothesis is supported for hippocampal tasks by direct observations of changes in cerebral glucose in rats, it is not clear whether such sensitivity exists for frontal lobe tasks (McNay, Fries, & Gold, 2000). Moreover, delivery of glucose from blood to active neurones is normally exquisitely buffered by astrocytes, to allow supply to meet demand (Peters et al, 2004), The lack of glucose facilitafion effects with higher doses (Smith et al, 2011) is also hard to explain in terms of repletion of glucose during demanding tasks, although such doses may provoke substantial cortisol release that is associated with memory impairment (Gibson, 2007), Finally, the present results demonstrated that glucose facilitafion effects can be observed for nonmedial temporal lobe tasks. Most research demonstrating glucose enhancement effects have done so using episodic memory tasks mediated by different areas of the
GLUCOSE AND STROOP medial temporal lobes such as the hippocampus (Riby et al., 2008; Smith et al., 2011; Sünram-Lea, Dewhurst & Foster, 2008) or the parahippocampus (Stone, Thermenos, Tarbox, Poldrack & Seidman, 2005). However, there is some demonstration that the beneficial effects of glucose administration can extend beyond the medial temporal lobes. For example, research has shown a trend toward enhanced activation after glucose administration in the left dorsolateral prefrontal cortex in schizophrenics (Stone et al., 2005). In addition, Riby et al. (2008) found some evidence that the P3a ERP frontal lobe component was enhanced after glucose administration in normal young adults, although they contended that variability in their data and a small sample size meant further work in this area is required. Finally Gagnon, Greenwood, and Bherer (2010) showed enhanced Stroop performance in 60- to 80-year-old adults, albeit with twice the dose of glucose used in the present research. To our knowledge, therefore, our results provide the first support for a reliable glucose facilitation effect in frontal lobe function in young adults and at the same low dose that appears to be most effective in enhancing medial temporal lobe function. One clear task for future research would be to investigate whether other frontal lobe tasks benefit from glucose ingestion and whether such benefits are of similar magnitude to those enjoyed by medial temporal lobes tasks.
References Awad, N., Gagnon, M., Desrochers, A., Tsiakas, M., & Messier, C. (2002). Impact of peripheral glucoregulation on memory. Behavioral Neuroscience, 116, 691-702. doi:10.1037/0735-7044.116.4.691 Benton, D., Owens, S., & Parker, P. Y. (1994). Blood glucose influences memory and attention in young adults. Neuropsychologia, 32, 595-607. doi: 10.1016/0028-3932(94)90147-3 Brandt, K. R., SUnram-Lea, S., Jenkinson, P. M., & Jones, E. (2010). Glucose and emotional memory: Moderating effects of dose and task difficulty. Behavioural Brain Research, 211, 83-88. doi:10.1016/j.bbr .2010.03.016 Donohoe, R. T., & Benton, D. (1999). Cognitive functioning is susceptible to the level of blood glucose. Psychopharmacology, 145, 378-385. doi:10.1007/s002130051071 Gagnon, C , Greenwood, C. E., & Bherer, L. (2010). The acute effects of glucose ingestion on attentional control in fasting healthy older adults. Psychopharmacoiogy, 211, 337-346. doi:10.1007/s00213-010-1905-9 Gagnon, C , Greenwood, C. E., & Bherer, L. (2011). Glucose regulation is associated with attentional control performances in nondiabetic older adults. Joumai of Clinical and Experimental Neuropsychology, 33, 972-981. doi:10.1080/13803395.2011.589372 Gibson, E. L. (2007). Carbohydrates and mental function: Feeding or impeding the brain? Nutrition Bulletin, 32, 71-83. doi:10.1 lll/j.14673010.2007.00606.x Green, M. W., Taylor, M. A., Elliman, N. A., & Rhodes, O. (2001). Placebo expectancy effects in the relationship between glucose and cognition. British Journal of Nutrition, 86, 173-179. doi:10.1079/ BJN2001398
935
Lamport, D. J., Lawton, C. L., Mansfield, M. W., & Dye, L. (2009). Impairments in glucose tolerance can have a negative impact on cognitive function: A systematic research review. Neuroscience and Biobeiiavioral Reviews, 33, 394-413. doi:10.1016/j.neubiorev.2008.10.008 McNay, E. C, Fries, T. M., & Gold, P. E. (2000). Decreases in rat extracellular hippocampal glucose concentration associated with cognitive demand during a spatial task. Proceedings of the National Academy of Sciences of the United States of America, 97, 2881-2885. doi: 10.1073/pnas.050583697 Messier, C. (2004). Glucose and memory, a review. European Journai of Pharmacology, 490, 33-57. doi:10.1016/j.ejphar.2004.02.043 Messier, C , Awad-Shimoon, N., Gagnon, M., Desrochers, A., & Tsiakas, M. (2011). Glucose regulation is associated with cognitive performance in young nondiabetic adults. Behavioural Brain Research, 222, 81-88. doi:10.1016/j.bbr.2011.03.023 Peters, A., Schweiger, U., Pellerin, L., Hubold, C , Oltmanns, K. M., Conrad, M., . . . Fehm, H. L. (2004). The selfish brain: Competition for energy resources. Neuroscience and Biobehaviorai Reviews, 28, 143180. doi:10.1016/j.neubiorev.2004.03.002 Riby, L., McLaughlin, J., & Riby, D. (2008). Lifestyle, glucose regulation and the cognitive effects of glucose load in middle-aged adults. The British Journal of Nutrition, 100, 1128-1134. Riby, L. M., Sünram-Lea, S. L, Graham, C , Foster, J. K., Cooper, T., Moodie, C , & Gunn, V. P. (2008). P3b versus P3a: An event-related potential investigation of the glucose facilitation effect. Joumal of Psychopharmacoiogy, 22, 486-492. doi:10.1177/0269881107081561 Scholey, A. B., Laing, S., & Kennedy, D. O. (2006). Blood glucose changes and memory: Effects of manipulating emotionality and mental effort. Biologicai Psychoiogy, 71, 12-19. doi:10.1016/j.biopsycho.2005 .02.003 Smith, M. A., Riby, L. M., van Eekelen, J. A. M., & Foster, J. K. (2011). Glucose enhancement of human memory: A comprehensive research review of the glucose memory facilitation effect. Neuroscience and Biobehaviorai Reviews, 35, 770-783. doi:10.1016/j.neubiorev.2010.09 .008 Stone, W. S., Thermenos, H. W., Tarbox, S. I., Poldrack, R. A., & Seidman, L. J. (2005). Medial temporal and prefrontal lobe activation during verbal encoding following glucose ingestion in schizophrenia: A pilot fMRI study. Neurobiology of learning and Memory, 83, 54-64. doi:10.1016/j.nlm.2004.07.009 Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18, 643-662. doi:10.1037/ h0054651 Sünram-Lea, S. I.. Dewhurst, S. A., & Foster, J. K. (2008). The effect of gluco.se administration on the recollection and familiarity components of recognition memory. Biological Psychology, 77, 69-75. doi:10.1016/j .biopsycho.2007.09.006 Sünram-Lea, S. I., Foster, J. K., Durlach, P., & Perez, C. (2002). Investigation into the significance of task difficulty and divided allocation of resources on the glucose memory facilitation effect. Psychopharmacology, 160, 387-397. doi:10.1007/s00213-001-0987-9 Received July 22, 2013 Revision received October 3, 2013 Accepted October 5, 2013 •
Copyright of Behavioral Neuroscience is the property of American Psychological Association and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
