Data were preprocessed with R software (38) and analysed with JASP 0.9.2 (39). RTs from the PVT and the category semantic priming task were log-transformed to reduce skewness in the distribution.
Regarding the PVT, the first trial, trials with RTs shorter than 150 ms, and trials with RTs separated by more than 3.5 semi-interquartile ranges from the median value of each participant in each session were considered outliers and discarded (1.77% of trials).
In relation to the category semantic priming task, we applied two different filters considering the graphical distribution of the RT raw data. In light of the graphical representation, we first established a long-time interval where most of the participants’ responses were concentrated: 100 ms to 4000 ms. Specifically, only 3 participants made some responses above 4000 ms, and no one made responses below 100 ms. Second, we considered outliers all RTs that were separated by more than 3.5 semi-interquartile ranges from the median value of each participant in each condition. The percentage of discarded trials did not exceed 1.10% in any experimental condition. Finally, mean RTs were calculated considering only RTs associated with correct responses and after discarding practice trials.
Analyses were performed on logarithmically transformed RTs, although the figures are plotted in RTs. For the accuracy data the statistical analysis was conducted with the proportion of correct responses. We adopted a statistical significance level of 𝛂 = .05 for all statistical analyses. Moreover, we present effect size values for each of our contrasts through the partial eta squared (ηp2) in ANOVA and Cohen's d (d) in Student’s t tests.
Demographic data
The scores on the rMEQ used to classify individuals according to their chronotype showed statistically significant differences among the three groups, F(2, 62) = 157.40; p < .001; ηp2 = .84. The analysis of age showed no statistically significant differences between the groups (p = .99). The participants reported 5 or more hours of sleep the night before the experiment. As instructed, none of the participants reported caffeine consumption in the two hours prior to the experiment.
Experiment 1: Intermediate-type Chronotype
The results of the analysis of the RTs are shown in Fig. 1a (short SOA) and Fig. 1b (long SOAs). With the 100-ms SOA, the mean log-RTs analysis produced no priming effects as the main effect of relatedness was not significant, t(23) = 1.51, p = .14, d = .31. However, an inspection to Fig. 1a reveals that the lack of priming may be due to the facilitatory effect observed in the first subblock of trials, which proved statistically significant (53 ms), t(23) = 2.54, p < .01, d = .51, being counteracted by the lack of effect in the second subblock (2 ms) (p = .78), and the non-significant trend for an inhibitory effect in the third subblock (-7 ms) (p = .96). With the longer SOAs, the mean log-RTs were subjected to a 3 × 2 repeated-measures ANOVA with prime-target SOA (450, 650 and 850 ms) and relatedness (related and unrelated) as within-participants factors. The main effects of SOA and relatedness were significant. On the one hand, the effect of SOA, F(2, 46) = 8.11, p < .001, ηp2 = .26, reflected RTs that decreased as SOA increased (691, 669, and 667 ms, for the 450-, 650- and 850-ms SOAs, respectively), and the relatedness effect, F(1, 23) = 10.87, p = .003, ηp2 = .32, showed that, in general, the participants were faster in unrelated (656 ms) than in related trials (695 ms). In addition, a significant SOA × relatedness interaction was also found, F(2, 46) = 10.11, p < .001, ηp2 = .30, which reflected a change in the priming effect across the SOAs. Further analyses of the interaction showed the progressive development of an inhibitory priming effect. With a SOA of 450 ms, priming (-9 ms) was not yet significant, t(23) = 1.57, p = .131, d = .32, but priming was statistically significant with SOAs of 650 ms (-35 ms), t(23) = 2.89, p = .008, d = .59, and 850 ms (-75 ms), t(23) = 4.03, p < .001, d = .82.
The results of the analysis of the accuracy data are shown in Fig. 1c. With the 100-ms SOA, the main effect of relatedness was not significant, t(23) = .15, p = .88, d = .03. Also, priming effects did not vary across the three subblocks of trials (all ps > .24). With the long SOAs, we observed a significant main effect of relatedness, F(1, 25) = 6.75, p = .01, ηp2 = .21, indicating higher proportion of correct responses in unrelated (.94) than in related trials (.90). No other effects or interactions were significant (all ps > .61).
Insert Fig. 1 here
An inspection to Fig. 1a and Fig. 1b suggests that intermediate-type participants showed automatic processing at the very beginning of the experimental session, being apparent only in the first subblock of the short prime-target SOA block of trials. The lack of priming effects in both the two later subblocks of the short SOA block and the 450 ms SOA of the long SOAs block, suggest that automatic and controlled processes cancelled out each other during that interval of time. With longer prime-target SOAs, automatic processing may have already dissipated and therefore the results show the development of controlled processing, which makes stronger with time. This interplay between automatic and controlled processing as a function of prime-target SOA suggests that the pattern of priming effects showed by intermediate-type participants constitutes an appropriate referent to assess the performance of extreme chronotype participants when testing occurs in their optimal and non-optimal time-of-day.
Experiment 2: Extreme Chronotypes
Psychomotor Vigilance Task (PVT)
Mean RTs were subjected to a mixed ANOVA with time-of-day (optimal and non-optimal) as the within-participants factor and chronotype (morning-types and evening-types) as the between-participants factor. The main effect of time-of-day was statistically significant, F(1, 38) = 5.61; p = .02; ηp2 = .13, which indicated that, in general, performance at the optimal time produced shorter RTs (319 ms) than at the non-optimal time (341 ms). The difference in performance between the optimal and non-optimal time-of-day according to the different chronotypes is referred to as the synchrony effect and the current results replicate those obtained in other experiments using the same task (11, 35, 40). Accordingly, the current synchrony effect confirms the appropriate selection of our sample of extreme chronotypes. No other effects or interactions were significant (all ps > .31).
Category Semantic Priming Task
The mean RTs and proportion of correct responses of the three chronotypes in all conditions of the category semantic priming task are presented as additional material [see category semantic priming data.xlsx].
Automatic and controlled processes in morning-type chronotype
The results of the RTs analysis are shown in Fig. 2a. With the 100-ms SOA, mean log-RTs were subjected to a 2 × 2 repeated-measures ANOVA with relatedness (related and unrelated) and time-of-day (optimal and non-optimal) as within-participants. The statistical analysis showed no significant main effects of relatedness, F(1, 19) = .24; p = .63; ηp2 = .01, time-of-day, F(1, 19) = 1.19; p = .29; ηp2 = .06, nor the interaction between the two factors, F(1, 19) = .12; p = .73; ηp2 = .006. In addition, in contrast to intermediate-type participants, who showed facilitatory priming just in the first subblock of trials with the 100-ms SOA, morning-type participants did not show any relatedness effect in any subblock of trials (all ps > .22). With the long SOAs, mean log-RTs were subjected to a 3 × 2 × 2 repeated-measures ANOVA with SOA (450, 650, and 850 ms), relatedness (related and unrelated) and time-of-day (optimal and non-optimal) as within-participants factors. The main effect of SOA was statistically significant, F(2, 38) = 15.84; p < .001; ηp2 = .45, which reflected a general decrease in RTs as SOA increased (733, 707, and 694 ms, for the 450-, 650- and 850-ms SOAs, respectively). In addition, we observed that RTs were longer for related (750 ms) than for unrelated trials (711 ms), F(1, 19) = 10.14; p = .005; ηp2 = .35, and this finding is compatible with an inhibitory priming effect. However, the SOA × relatedness × time-of-day interaction did not prove to be statistically significant, F(2, 38) = 2.47; p = .09; ηp2 = .11. Thus, we did not find any modulation of priming due to time of testing for the morning-type participants in the analysis of RTs.
The results of the accuracy analysis are shown in Fig. 2b. With the 100-ms SOA, we observed no main effect of relatedness and time-of-day (all ps > .30). With the long SOAs, we observed a significant main effect of relatedness, F(1, 19) = 6.90; p = .02; ηp2 = .27, that is the proportion of correct responses was higher in unrelated (.98) than in related trials (.94). Moreover, the SOA × relatedness × time-of-day interaction was statistically significant, F(2, 38) = 4.67; p = .01; ηp2 = .20. The interaction was due to a significant synchrony effect (the difference in priming between the optimal and non-optimal time-of-day) only with the 850-ms SOA, t(19) = 2.79, p = .01, d = .62).
Insert Fig. 2 here
Automatic and controlled processes in evening-type chronotype
The results of the RTs analysis are shown in Fig. 3a. With the 100-ms SOA, mean log-RTs were subjected to a 2 × 2 repeated-measures ANOVA with relatedness (related and unrelated) and time-of-day (optimal and non-optimal) as within-participants. Only the relatedness × time-of-day interaction was significant, F(1, 19) = 9.28; p = .007; ηp2 = .34. The analysis of the interaction showed that facilitatory priming (16 ms) was observed only at the non-optimal time of day, t(19) = 2.14, p = .04, d = .48), and inhibitory priming (-29 ms) was observed only at the optimal time of day, although the effect was marginally significant, t(19) = 1.91, p = .07, d = .42. Both priming effects differed significantly, t(19) = 3.05, p = .007, d = .68. In line with the morning-type participants, we did not observe any priming difference across the three subblocks of trials (all ps > .23). With the long SOAs, mean log-RTs were subjected to a 3 × 2 × 2 repeated-measures ANOVA with SOA (450, 650, and 850 ms), relatedness (related and unrelated) and time-of-day (optimal and non-optimal) as within-participants factors. We observed significant main effects of SOA, F(2, 38) = 16.96; p < .001; ηp2 = .47, relatedness, F(1, 19) = 12.64; p = .002; ηp2 = .40, and time-of-day, F(1, 19) = 4.95; p = .03; ηp2 = .20. RTs decreased with SOA (656, 639, and 623 ms, for the 450- 650- and 850-ms SOAs, respectively), were larger for related (651 ms) than for unrelated trials (628 ms), and for the non-optimal (667 ms) than for the optimal time-of-day (611 ms). In addition, three interaction effects were statistically significant: relatedness × time-of-day, F(1, 19) = 6.43; p = .02; ηp2 = .25, SOA × relatedness, F(2, 38) = 3.15; p = .05; ηp2 = .14, and SOA × relatedness × time-of-day, F(2, 38) = 4.97; p = .01; ηp2 = .20. The analysis of the three-way interaction showed significant inhibitory priming at the 600-ms SOA, t(19) = 5.16, p < .001, d = .57, and 850-ms SOA, t(19) = 3.47, p = .003, d = .77, when testing occurred at the optimal time-of-day, and only at the 850-ms SOA, t(19) = 2.30, p = .03, d = .51, when testing occurred at the non-optimal time-of-day.
The results of the accuracy analysis are shown in Fig. 3b. With the 100-ms SOA, we observed a main effect of relatedness, F(1, 19) = 7.22; p = .01; ηp2 = .28, indicating higher proportion of correct responses in related (.97) than in unrelated trials (.95), reflecting automatic facilitatory priming. With the long SOAs, we also observed a main effect of relatedness, F(1, 19) = 7.91; p = .01; ηp2 = .29, indicating higher proportion of correct responses in unrelated (.96) than in related trials (.91). There were no other statistically significant main effects nor interactions (all ps > .14).
Insert Fig. 3 here