*2.2. Results*

As presented in Table 1, there were no differences between the two groups of restrained eaters in age, BMI, and DEBQ-R. Before assessing differences in food consumption between the groups, we conducted a t-test to assess differences in hunger level between the groups before the training and found no differences in hunger level between the food-response training group (mean = 4.47, SD = 2.23) and the food-response/inhibition training group (mean = 3.97, SD = 1.77) (*t*(62) = 0.99, *p* = 0.324).

To ensure proper task engagement, we calculated two measures for the F-SST task: the nsRT (RT in no-stop-signal trials) and the nsACC (accuracy in no-stop-signal trials). There were no significant differences between the groups in both nsRT (food-response training: mean = 540, food-response/ inhibition training: mean = 527; *t*(62) = 0.71, *p* = 0.480) and nsACC (food-response training: mean = 0.96, food-response/inhibition training: mean = 0.95; *t*(62) = 0.55, *p* = 0.585), indicating similar task engagement in both groups, on measures that are not affected by the specific version of the task.

Next, food consumption was calculated for each participant as the sum of snacks tasted from all three bowls. To test our primary hypothesis, independent t-tests were carried out to assess differences between the training groups in food consumption. As was hypothesized, the results showed reduced food consumption in the food-response/inhibition training group compared to the food-response training group. This was evident both in number of snacks eaten (5.53 (SE = 0.39) in the food-response group vs. 4.34 (SE = 0.29) in the food-response/inhibition group*; t*(62) = 2.45, *p* = 0.017, Cohen's d = 0.61; Figure 2A) and in the total amount of kcal consumed (48.88 (SE = 4.23) in the food-response group vs. 37.34 (SE = 3.27) in the food-response/inhibition group; *t*(62) = 2.16, *p* = 0.034, Cohen's d = 0.54).

**Figure 2.** (**A**) Differences in snack consumption between the food-response and the food-response/ inhibition training groups. The *y-*axis represents the total number of snacks eaten from the three bowls. (**B**) Changes in food-related anxiety as a function of training group and time. The *y*-axis shows the food-related anxiety score on the visual analog scale (VAS). The *x*-axis represents group. Error bars represent 1 standard error from the mean. \* indicates *p* < 0.05.

In order to assess changes in food-related anxiety following the trainings, we conducted a two-way mixed model (ANOVA) on food-related anxiety score on the VAS with time (pre vs. post) as a within-subject factor and group (food-response training vs. food-response/inhibition training) as a between-subject factor. There were no main effects for time (*F*(1, 62) = 0.37, *p* = 0.547) or group (*F*(1, 62) = 2.08, *p* = 0.154). Importantly, the group × time interaction was significant (*F*(1, 62) = 5.46, *p* = 0.023, η<sup>p</sup> <sup>2</sup> = 0.08). Planned comparisons revealed a trend toward increased food-related anxiety from pre- to post-training in the food-response training group (*t*(31) = 1.91, *p* = 0.06, Cohen's d = 0.56; Figure 2B), whilst in the food-response/inhibition group, there was a reduction in food-related anxiety from pre- to post-training, but this did not reach significance level (*t*(31) = 1.36, *p* = 0.18; Figure 2B).
