Caffeine and Cognition Write-Up 1: Which data sets have you selected for your analysis? (5% of mark) The question being considered is the “battle of the sexes” and whether males or females are better at speechbased tasks (this study used the Stroop test). The second question is whether or not caffeine improved the performance of participants in the speech based tasks. Data sets used: V1(Sex of participant), V3(What was the drink taken), V23 (Stroop test-time taken to read word) and V24 (Stroop test-time taken to read colour) 2: What is the background of your study? (25% of mark) Caffeine is one of the most commonly used CNS stimulants (1).
Cytochrome P450 oxidase metabolises caffeine to Paraxanthine, Theobromine and Theophylline (2). Caffeine acts as a competitive inhibitor of adenosine receptors (3); Adenosine and adenosine receptors play a role in regulating sleep. The levels of adenosine rise with more wakefulness and therefore increased binding to adenosine receptors (4). This binding causes tiredness. Caffeine is an inhibitor of all adenosine receptors; however its main effects are achieved through inhibition of the A1 receptor. The action of caffeine is to antagonise A1 receptors preventing calcium channel blockage thus depolarizing the neuronal cell.
This occurs as the presynaptic A1 receptor is G-protein coupled and binding of caffeine blocks the action of the G-protein; thereby adenylate cyclase is not inhibited, which leads to potassium channels closing and calcium channels opening (5). The A1 adenosine receptor is found in all parts of the brain from the hippocampus to the cortex (6). The part of the brain important for speechbased tasks is called the Broca’s area. When significantly damaged the speech ability of an individual is greatly reduced (7).
Therefore the neuronal activation of the Broca’s area will be important for participating in the Stroop test. 10% 3: What is your hypothesis? (10% of mark) Hypothesis 1: No clear differences will be evident when male and female results for both parts of the Stroop test are compared. Hypothesis 2: Caffeine will have no effect on the time taken to complete both parts of the Stroop test. 4: What are your results? (30% of mark) Hypothesis 1 To test hypothesis 1, two boxplots were created with each displaying one part of the Stroop test. Part 1: Figure 1: Boxplot of Stroop test comparing male and female times to read the word and not the colour. The number of participants stayed consistent throughout the experiments.
Since there was roughly the same number of participants with caffeine and placebo, a comparison was not shown on the graphs. The result for part 1 showed there was no difference in the mean results. An ancova was carried out to see if the result was statistically significant. To do this a linear model was first created using R studio (V23 = Read word Stroop test, V1 = Sex): model1F) V3 122 22. 00 1. 328 0. 254 Residuals 56 928 16. 57 —- Table 1: The results of the ancova for comparing reading the word Stroop test and Sex.
The important result is the P value. It indicates significance as it s the percentage probability of the results occurring by chance. For results to be statistically significant, and therefore a high chance that if the experiment was repeated the same results would be acquired, the P value would have to be 0. 05 or less. It is observed to be 0. 254, therefore this result could occur in 1 in every 2. 54 repeats by chance, meaning the results were not significant. Part 2: Figure 2: Boxplot of Stroop test comparing male and female times to read the colour and not the word. There was no observable difference in mean times to complete the task.
An ancova was conducted using the same method as before except V24 (Read color Stroop test) time was used instead of V23: model2F) V3 188. 9 88. 94 3. 516 0. 066 Residuals 56 1416. 5 25. 29 ———— Table 2: The results of the ancova for comparing reading the colour Stroop test and Sex. The P value, 0. 066 was lower this time but still not significant as it was greater than 0. 05. Hypothesis 2 To test hypothesis 2, the same methods were used as to test hypothesis 1. Male and female categories were not considered for this part as only the effect of caffeine was being tested.
Part 1: Figure 3: Boxplot of Stroop test comparing caffeine and placebo times to read the word and not the colour. Figure 3 showed that participants that had caffeine also had a slightly lesser mean time suggesting caffeine had a partial effect. To test whether this result was significant or not, an ancova test was conducted. This time using V23 but V3 (Caffeine or placebo) instead of V1: model3F) V3 17. 17. 124 0. 423 0. 518 Residuals 56 942. 8 16. 836
Table 3: The results of the ancova for comparing reading the word Stroop test and caffeine or placebo. The P value observed is 0. 518 which is higher than 0. 5; therefore the result was insignificant. Part 2: Figure 4: Boxplot of Stroop test comparing caffeine and placebo times to read the colour and not the word. Figure 4 shows a slight decrease in the mean time taken to complete the task in the caffeine group. This was tested for significance using an ancova test with V23 changing to V24 but still using V3: model4F) V3 1 20. 5 20. 54 0. 775 0. 383 Residuals 56 1484. 9 26. 52 Table 1: The results of the ancova for comparing reading the colour Stroop test and Caffeine and placebo. The P value was 0. 383 and 0. 05 or less so the result is insignificant. : What do you conclude? (20% of mark)
The results for testing hypothesis 1 would suggest that there is no clear difference between the mean time taken by females and the mean time taken by males for the Stroop test. This shows that in the “battle of the sexes”, neither sex outperformed the other. For hypothesis 2, the results indicated that caffeine had a partial effect on the mean time to complete the Stroop test; participants that had ingested caffeine had a slight decrease in time compared to the placebo. This decrease may be due to the increased depolarisation in the Broca’s area.
The binding of caffeine to A1 receptors may allow better use of the speech mechanisms. The competitive binding of caffeine blocks the binding of adenosine to A1 receptors. This halts the signaling cascade that leads to tiredness; therefore the participants with caffeine would feel less tired and perform better, so this may also explain the decrease. However, testing of the linear models in all 4 parts of the experiment, using an ancova test, revealed that all results were statistically insignificant. This means that the results cannot be accepted and the hypothesis proved or disproven.
Therefore the design of this test would need to be improved to obtain statistically significant data. An improvement would be to check how regularly participants consumed caffeine; as tonic usage can cause tolerance of caffeine and therefore no effect would be seen (8). A power analysis would also improve the study as it would show the correct number of participants needed to make statistically significant results. A final improvement would be to use a larger dose of caffeine to see whether increased caffeine caused a notable effect on the experiments.
