Demographic information

A total sample of 298 participants were included in this study. The MDD group included 168 patients, with age ranged from 18 to 58 years (M = 35.34, SD = 9.53), of which 57.1% were female. The HC group included 130 healthy participants, with age ranged between 18 and 45 years (M = 32.08, SD = 7.76), of which 56.2% were male. Within the MDD group, the average of onset age of depression was 31.94 years, and the average number of episodes of depression was 2.08. The average scores of HAMD24, HAMA, and C-DAS-A total of MDD group were much higher than that of HC group, whereas the average years of education of MDD patients were lower than healthy participants. There were statistically significant differences in all these variables between MDD and HC groups (p < .01).

Almost all of the 16 PF factors were significantly different (p < .05) between the MDD group and HC group, except for the factors A, I, M, N, Q1, and Q2. As for the anxiety facets of 16 PF, the mean scores of C factor of the MDD group were lower than that of the HC group, whereas the mean scores of L, O, and Q4 factors of the MDD group were higher than the HC group, indicating an overall higher level of anxiety in participants with MDD. Table 1 presents demographic and clinical characteristics of MDD and HC groups.

Table 1 Means and standard deviations of major study variables of the MDD group and the HC groupCorrelations of anxiety facets of 16PF and dysfunctional attitudes and subtypes of MDD and HC groups

Pearson’s correlations demonstrated significant associations between specific domains of 16 PF and dysfunctional attitudes in both MDD group and HC group (see Tables 2 and 3). Cohen’s (1992) standards for Pearson’s correlation coefficient effect size were used to determine the strength of the effects (i.e., small, 0 ≤ r < .3; medium, 0.3 ≤ r < .5; large, 0.5 ≤ r ≤ 1) [5]. As expected, the 16PF dimensions related to the dysfunctional attitudes were mainly concentrated in the four anxiety facets of 16 PF. Within the MDD group, specifically, C factor showed a medium negative correlation with dysfunctional attitudes (r = − .35, p < .001), whereas other three primary personality traits were positively associated with dysfunctional attitudes with small to medium correlation: L factor (r = .16, p = .036), O factor (r = .40, p < .001), and Q4 factors (r = .30, p < .001).

Table 2 Pearson’s correlation coefficients of 16PF and dysfunctional attitudes and subtypes in the MDD group (n = 168)Table 3 Pearson’s correlation coefficients of 16PF and dysfunctional attitudes and subtypes in the HC group (n = 130).

Within the HC group, C factor were negatively correlated with dysfunctional attitudes with a medium correlation (r = − .34, p < .001), the L factor (r = .33, p < .001) and O factor (r = .41, p < .001) were positively correlated with dysfunctional attitudes in a medium range, and Q4 factor (r = .51, p < .001) showed a strong positive correlation with dysfunctional attitudes.

Hierarchical regression analyses of 16 PF on dysfunctional attitudes and subtypes of MDD and HC groups

Table 4 shows the results of the hierarchical regression analyses of 16 PF on dysfunctional attitudes and subtypes in the MDD group. The 16PF dimensions which predicted dysfunctional attitudes score were C factor (β = − 0.19), G factor (β = − 0.16), and O factor (β = 0.26), with a ΔR2 of 21%. Additionally, the predictive effect of certain 16PF dimension on certain DAS subtype was found. Specifically, G factor and O factor would predict attraction and repulsion, C factor would predict perfectionism, B factor would predict compulsion, C factor and G factor would predict seeking applause, C factor and O factor would predict dependence, and G factor and O factor would predict cognition philosophy.

Table 4 Hierarchical regression analyses of 16PF on dysfunctional attitudes and subtypes in the MDD group (n = 168)

The results of the hierarchical regression analysis of the HC group were displayed in Table 5 The 16PF dimensions which predicted dysfunctional attitudes were L factor (β = 0.19) and Q4 factor (β = 0.44), with a ΔR2 of 25%. The predictive effects of certain 16PF dimension on certain DAS subtype were as follows: C factor, L factor, and Q4 factor would predict vulnerability; L factor, N factor, Q3 factor, and Q4 factor would predict attraction and repulsion; Q4 factor would predict compulsion; Q2 factor and Q4 factor would predict compulsion; L factor and Q4 factor would predict dependence; E factor and Q4 factor would predict self-determination attitude; and O factor would predict cognition philosophy.

Table 5 Hierarchical regression analyses of 16 PF on dysfunctional attitudes and subtypes in the healthy control group (n = 130)