Is dream text length a proxy for dream length?

Abstract

This is the result of a 34-month long longitudinal study of the author’s dreams.  The author has an unusually high dream recall frequency (average 45/month for 32 years), and during the period studied, from late June 2021 through April 2024, this increased to an average of over 75/month, for a total of 2,765 dreams over the course of 1,110 nights.  With some observed exceptions, the analysis showed that the percentage of nights with dream texts being consistently longer over the course of the night hovered around 70%, depending upon which measure was being used.  Two mitigating factors which offset this trend are discussed.  This tends to confirm findings from dream lab studies with much smaller N that suggest that dream text length does correlate – albeit loosely – with dream length.  To test the integrity of this data set, two random 1% samples of the ca. 16,000 dreams in the author’s dream database which were dreamt prior to the inception of the project were taken – before the author had formulated the idea that dream text length might be related to dream position and might be a proxy for dream length.  The results were slightly less definitive than for the dreams from the study period, but once the mitigating factors were taken into account, the majority of dreams in the random sample did follow the pattern, suggesting that it is valid, at least for the author’s dreaming.