Considering first the total recall of all slides in the emotional and neutral stories, a two-way ANOVA with emotional story content and hormonal contraception status (NC versus HC) as independent factors revealed that both NC (n = 32) and HC (n = 34) women recalled significantly more slides from the emotional than neutral stories (F(1,63) = 12.6, P = .0007, see ). The main effect of contraceptive status was non-significant (F(1, 63) = 1.38, P = .24) as was the interaction effect of emotion × contraceptive status (F(1, 63) = .06, P = .81). As predicted on the basis of earlier work with this story paradigm, the enhancement of total slide recall was driven by enhanced slide recall in the emotional component of the emotional story, phase 2. A two-way ANOVA with emotion and contraception status as independent factors showed that both NC and HC women recalled significantly more phase 2 slides from the emotional compared with neutral stories. (F(1, 63) = 18.6, P < .0001). The main effect of HC was non-significant (F(1, 63) = .8441, P = .36) as was the interaction effect of emotion × HC (F(1, 63) = 1.15, P = .29). The same analysis for phases 1 and 3 revealed no significant main effects or interactions.
Total slide recall in NC and HC women. A two-way ANOVA revealed a main effect of emotion on total slide recall in both NC and HC women (n = 32 NC women, n = 34 HC women; three asterisks, P<.001). Values are means ± s.e.m.
Although NC and HC women displayed equivalent total slide recall, they exhibited substantially different memory for gist and detail information from the emotional story. NC women (n=16 per condition) recalled significantly more details (F(1, 30) = 7.1, P = .012) from the emotional (M = 7.25, SD = 4.82) compared with neutral (M = 3.56, SD = 2.71) story. However, NC women recalled the gist no differently (F(1, 30) = .039, P = .85, see ) in the emotional (M = 3.56, SD = 1.89) as compared to the neutral (M = 3.38, SD = 3.30) condition.
Fig. 2 Total gist and detail item recall in NC and HC women. a, NC women exhibited an effect of emotion for detail retention but not for gist (n = 32 NC women; asterisk, P<.05, one-way ANOVA). Values are means ± s.e.m. b, HC women exhibited an (more ...)
In contrast, HC women (n=17 per condition) exhibited no enhancement of memory for details (F(1, 32) = 1.8, P =.19) in the emotional (M = 5.76, SD = 2.77) versus the neutral (M = 4.53, SD = 2.58) conditions, but showed a significant enhancement of memory for gist (F(1, 32) = 7.83, P = .008, see ) from the emotional (M = 5.94, SD = 2.95) story compared to the neutral (M = 3.18, SD = 2.93).
We next examined which story phase was driving the overall enhancements for gist in the HC women, and detail in the NC women. For NC women (n = 16 per cell), the emotional enhancement for detail memory was driven by recall from the emotionally-charged phase 2 (F(1, 30) = 17.9, P =.0002, see ). No significant enhancements of detail recall were seen for phases 1 and 3, nor were any significant enhancements for recall of gist seen in NC women in any story phase (). For HC women (n = 17 per cell), emotionally enhanced recall of gist was driven by phase 1 (F(1,32) = 9.50, P = .0042) and phase 2 (F(1, 32) = 4.86, P = .0347, see ). No difference in recall of gist between the emotional versus neutral stories was detected for phase 3, nor was any difference in recall of story details detected for HC women in any story phase ().
Fig. 3 Recall of detail and gist items by story phase in NC and HC women. a, NC women exhibited an effect of emotion on detail retention in Phase 2 (n = 32; two asterisks, P<.01, one-way ANOVA) whereas HC women did not (n = 34) exhibit an effect of emotion (more ...)
We next examined differences in gist and detail memory in HC and NC women specifically during phase 2 of the emotional story. A one-way ANOVA revealed that HC women (M = 2.18±.27, SD = 1.13) recalled significantly more gist items in phase 2 of the emotional story (F(1, 32) = 4.53, P =.04; see ) as compared to NC women (M = 1.38±.26, SD = 1.02). A second one-way ANOVA revealed that NC women (M = 3.75±.57, SD = 2.27) recalled significantly more detail items in phase 2 of the emotional story (F(1, 32) = 3.84, P =.05; see ) than did HC women (M = 2.41±.39, SD = 1.62).
Fig. 4 Recall of gist and detail items in phase 2 of the emotional story in NC and HC women. a, HC women (n = 17) recalled significantly more gist items (one asterisk, P = .05, one-way ANOVA) than NC women (n = 16) b, NC women (n = 16) recalled significantly (more ...)
Finally, to better understand the nature of the recall differences, we assessed the subjects’ scores on the AQ, TAS-20, and Mehrabian. Three separate one-way ANOVAs were used. NC women (M = 16.4 ± 0.87) did not differ significantly from HC women (M = 16.0 ± 0.97) on the AQ. On the TAS-20 and the Mehrabian, NC women (M = 42.8 ± 1.76 and M = 2.6 ± 1.9, respectively) and HC women (M = 43.6 ± 1.92 and M = 2.7 ± 1.7, respectively) also exhibited no significant differences.
Both attentional and arousal processes in the subjects during initial story viewing (assessed with eye-tracking technology) were very similar. As an index of attention, we determined the average fixation time for the four slides constituting phase 2, in which emotional elements occurred in the emotional story. NC (n
=16) and HC (n
=17) women spent equivalent percentages of time fixated on the slides in phase 2 of the emotional story (F(1, 31)
= 1.2, P
= .28, see ). In addition to fixation time, we examined whether the subjects explored the slides similarly in the emotional condition as indicated by the number of glances and subsequently, fixations, made in the specified AOIs (Poole & Ball, 2005
). No significant differences between NC (n
=16) and HC (n
=17) women were detected on either measure in any of the AOIs (data not shown).
Fig. 5 Attention and arousal in Phase 2 of the emotional story. a, No significant differences in average fixation time between NC and HC women (n = 16 and n = 17, respectively; P >> 0.1, one-way ANOVA). Values ± s.e.m. b, No significant (more ...)
Pupillary dilation indicated arousal to the emotional components (phase 2) of the emotional story (Koss, 1986
; Rajkowski, Kubiak, & Aston-Jones, 1993
; Aston-Jones & Cohen, 2005
). NC and HC women showed equivalent average pupil dilation in response to Phase 2 slides of the emotional story (F(1, 31)
= .057, P
= .81, see ). Collectively, the eye movement and dilation findings suggest that the differences in retention observed between NC and HC women cannot be explained by any potential effects of hormonal contraception either on the attentional focus of the women while they viewed the emotional story, or in the degree of arousal the story induced.