The observer’s blind spot: Discrepancies between self-reported and external ratings of dream affect

Abstract

Because dream experiences are largely accessible only through retrospective reports after awakening, estimates of dream affect may depend on who evaluates the report. We compared dreamers' self-ratings with ratings made by independent external judges to examine how rating method influences estimates of the presence, valence, and intensity of dream affect. The dataset comprised 1,946 home dream reports from 106 adults; models that included age used data from 104 participants. Dreams were judged as significantly more affective and emotionally intense when self-rated than when evaluated by external judges. External judges classified fewer dreams as containing affect and fewer dreams as positively valenced than did self-ratings. A three-way interaction among rating method, valence, and participants' biological sex indicated that biological sex influenced self-ratings but not external ones: women reported more intense negative dreams than men, whereas men rated their negative dreams as less intense than their own positive dreams; these differences disappeared when external ratings were used. While word count positively predicted affective occurrence and intensity, it did not account for valence. Self- and external ratings showed a significant positive association, although agreement at the level of individual dream reports was modest. Agreement was higher for negative than positive classifications. These results indicate that rating methods shape empirical portraits of dream affect and that self- and external ratings capture overlapping but noninterchangeable aspects of remembered dream experience.