Figure 1
The paradigm (a.) for Experiments 1 and 3, illustrating the prime and probe phase separated by either a task-switching task (Exp. 1) or a clock task (Exp. 3) that lasted for .5 (Exp. 1 only), 1, 3, 5, or 7 minutes (Exp. 1 & 3). For each task-switching task trial in these experiments, the stimulus is presented in the center of the screen with letters on either side indicating the classification task and response mapping. The prime and probe phases lasted for 8 trials each, respectively. The paradigm (b.) for Experiment 2 involved the continuous presentation of task-switching trials for 100 trials in each block, with no intervening filler task between prime and probe phases. Illustration (c.) of the manipulation of probe trials as either task switch or repeat trials, while keeping response mapping and completed task constant between the prime and probe presentation of an item (here, the example being the chair).
Table 1
Average number of per participant trials submitted to analyses, grouped by fixed factor levels, for Experiments 1, 2, and 3.
| PRIME TASK SWITCH | ||||||
|---|---|---|---|---|---|---|
| SWITCH | REPEAT | |||||
| PROBE TASK SWITCH | ||||||
| SWITCH | REPEAT | SWITCH | REPEAT | |||
| Exp 1 | Delay (min) | 0.5 | 3.1 | 2.6 | 2.8 | 2.8 |
| 1.0 | 3.2 | 2.9 | 2.9 | 3.1 | ||
| 3.0 | 3.1 | 2.7 | 2.8 | 3.0 | ||
| 5.0 | 3.1 | 2.7 | 2.7 | 3.0 | ||
| 7.0 | 2.9 | 2.9 | 2.6 | 3.1 | ||
| Exp 2 | Delay (trials) | 0–20 | 14.0 | 20.2 | 16.6 | 17.3 |
| 21–40 | 14.8 | 18.3 | 18.4 | 15.7 | ||
| 41–60 | 14.7 | 18.5 | 19.0 | 15.5 | ||
| 61–80 | 15.1 | 18.1 | 17.9 | 15.1 | ||
| 81–100 | 12.9 | 16.6 | 16.7 | 13.2 | ||
| Exp 3 | Delay (min) | 0.5 | 3.1 | 2.6 | 2.8 | 2.8 |
| 1.0 | 2.8 | 2.6 | 2.8 | 2.7 | ||
| 3.0 | 3.0 | 2.7 | 2.6 | 2.8 | ||
| 5.0 | 2.8 | 2.6 | 2.8 | 2.7 | ||
Table 2
Results of model comparison for hierarchical models of task-switching for Experiment 1.
| PARAMETERS | AIC | logLIK | CHI-SQUARED | df | p | ||
|---|---|---|---|---|---|---|---|
| 1. | Null | 4 | 177023 | -88507 | |||
| 2. | Probe Task Sequence | 5 | 176848 | -88419 | 176.83 | 1 | <.001 |
| 3. | + Prime Task Sequence | 6 | 176850 | -88419 | 0.23 | 1 | 0.630 |
| 4. | × Prime Task Sequence | 7 | 176852 | -88419 | 0.10 | 1 | 0.755 |
| 5. | + Delay | 8 | 176772 | -88378 | 81.17 | 1 | <.001 |
| 6. | × Delay | 11 | 176773 | -88376 | 5.00 | 3 | 0.172 |
Table 3
Summary results of the Probe task sequence × Prime task sequence + Delay model, in Experiment 1.
| ß | St.Err | t | p | |
|---|---|---|---|---|
| Intercept | 971.80 | 10.88 | 89.36 | <.001 |
| Probe Task Sequence | –48.58 | 4.94 | –9.84 | <.001 |
| Prime Task Sequence | –2.78 | 4.89 | –0.57 | 0.569 |
| Delay | 6.41 | 0.71 | 9.02 | <.001 |
| Probe × Prime Task Sequence | 2.26 | 6.95 | 0.33 | 0.745 |
Figure 2
Response times (ms ± 95% estimated confidence intervals) for probe trials, plotted as a function of the Probe Task Sequence (repeat vs. switch), the Prime Task Sequence (repeat vs. switch), and the Delay between prime and probe trials for Experiment 1.
Figure 3
Illustration of the shuffling process for Experiment 2 where primes and probes were submitted to a pseudo-randomized shuffle within 20-trial ‘bins’. The order of presented ‘bins’ during the prime stage was reversed in the probe stage (i.e. if bin order was 1, 2, 3, 4, 5 in the prime, it would be 5, 4, 3, 2, 1 in the probe stage).
Table 4
Results of model comparison for hierarchical models of task-switching for Experiment 2.
| PARAMETERS | AIC | logLIK | CHI-SQUARED | df | p | ||
|---|---|---|---|---|---|---|---|
| 1. | Null | 4 | 337365 | –168679 | |||
| 2. | Probe Task Sequence | 5 | 337278 | –168634 | 89.34 | 1 | <.001 |
| 3. | + Prime Task Sequence | 6 | 337274 | –168631 | 5.45 | 1 | 0.020 |
| 4. | × Prime Task Sequence | 7 | 337270 | –168628 | 6.06 | 1 | 0.014 |
| 5. | + Distance | 8 | 337221 | –168603 | 51.00 | 1 | <.001 |
| 6. | × Distance | 11 | 337224 | –168601 | 3.52 | 3 | 0.319 |
Table 5
Summary results of the Probe task sequence × Prime Task Sequence + Distance model in Experiment 2.
| ß | St.Err | t | p | |
|---|---|---|---|---|
| Intercept | 968.95 | 23.44 | 41.34 | <.001 |
| Probe Task Sequence | –34.43 | 4.06 | –8.48 | <.001 |
| Prime Task Sequence | –0.41 | 4.02 | –0.10 | 0.920 |
| Distance | 7.33 | 1.03 | 7.15 | <.001 |
| Probe × Prime Task Sequence | 14.63 | 5.77 | 2.54 | 0.011 |
Figure 4
The response times (ms ± 95% confidence intervals) for probe trials, plotted as a function of Probe Task Sequence (repeat vs. switch), Prime Task Sequence (repeat vs. switch), and Distance, the number of trials between the prime and probe presentation of an image, for Experiment 2. Estimated time between prime and probe is displayed below each distance.
Table 6
Results of model comparison for hierarchical models of task-switching for Experiment 3.
| PARAMETERS | AIC | logLIK | CHI-SQUARED | df | p | ||
|---|---|---|---|---|---|---|---|
| 1. | Null | 4 | 133654 | –66823 | |||
| 2. | Probe Task Sequence | 5 | 133518 | –66754 | 138.40 | 1 | <.001 |
| 3. | + Prime Task Sequence | 6 | 133519 | –66753 | 0.90 | 1 | .344 |
| 4. | × Prime Task Sequence | 7 | 133505 | –66745 | 16.02 | 1 | <.001 |
| 5. | + Delay | 8 | 133447 | –66716 | 59.37 | 1 | <.001 |
| 6. | × Delay | 11 | 133446 | –66712 | 6.83 | 3 | .078 |
Table 7
Summary results of the Probe task sequence × Prime Task Sequence + Delay model in Experiment 3.
| ß | St.Err | t | p | |
|---|---|---|---|---|
| Intercept | 937.42 | 13.09 | 301.92 | <.001 |
| Probe Task Sequence | –26.33 | 4.87 | –5.41 | <.001 |
| Prime Task Sequence | 10.29 | 4.76 | 2.16 | 0.031 |
| Delay | 5.86 | 0.76 | 7.72 | <.001 |
| Probe × Prime Task Sequence | –27.80 | 6.81 | –4.08 | <.001 |
Figure 5
Response times (ms ± 95% estimated confidence intervals) for probe trials, plotted as a function of Probe Task Sequence (repeat vs. switch), Prime Task Sequence (repeat vs. switch), and the Delay between prime and probe trials for Experiment 3.