No abstract

The production of stimulus-independent thought (e.g. fantasy and imagery) was measured as B function of the rate at which information was presented to human subjects. Information in the form of simple tones WES presented at rates from 0.2 to 6 bits per sec. The 1in-w~~ regression of reported stimulus-independent thought on information rate accounted for 83 per cent of the between-cell variance. The results support a. model in which both sensory and memory events are operated on by a common central cognitive unit.The intrusion of unsolicited thoughts, images or fantasy while engaged in reading, driving or listening to a lecture is an experience familiar to everyone-the stream of consciousness in William James's classical phrase (1890). Such thoughts may supplement the primary task of driving an automobile, or, when listening to a lecture, may altogether override one's intended goal. These illustrations emphasize the large fraction of one's cognitive activity that is independent of the concurrent external stimuli to which one is exposed. This cognitive activity corresponds, in the vernacular, to thought, imagery and fantasy, as distinct from sensations and perceptions. The distinction concerns the extent to which conscious cognitive activity 'follows ', or covaries in time, with stimulus events external to the cognitive system. Perceptual events are highly correlated; fantasy may be almost independent. The contrast is sharpened by an analogous distinction in computer systems. To the extent that the human cognitive system not only receives and stores but also carries out subsequent transformations on its information input, it behaves like a computer that operates sometimes on information as it arrives in 'real time', sometimes on information previously stored in memory, or simultaneously on both.The present experiment is one of a series concerned with the rules that determine when a cognitive system operates on sensory input-real time processing-and when it operates on events previously stored in memory. Using an information-processing approach t o the problem, both sensory-perceptual and non-perceptual events are assumed to be outputs of a common limited-capacity cognitive operator. It is further assumed that a pay-off value is associated with each perceptual or non-perceptual operation which the cognitive operator might make, and that the cognitive system operates to maximize the total pay-off of these operations. If the input from combined sensory and memory input presented to the cognitive system exceeds the capacity of the cognitive operator, the total pay-off may be maximized by gating out input . associated with low pay-off. Prior to a job interview, an examination or marriage, one may experience some difficulty in maintaining one's thoughts on lectures, reading and other tasks. The intruding, often repetitive thoughts may, as undeliberate planning, achieve their 'pay-off' in improved performance in subsequent overt behaviour . More typically, however, daydreaming and similar transformations on

The difference between Sleep Stages 1 REM (rapid eye movement) and NREM (non REM) in reported dreaming is compared to several word frequency measures of characteristics of sleep mentation reports. Total Recall Frequency (total word frequency minus pauses, fillers, corrections, repetitions and commentary) discriminated REM from NREM with approximately the same strength as did global ratings of Dreaming. Although incremental repeated measures multivariate tests showed that each contributed some unique variance to the REM/NREM criterion, the unique portions were trivial in size. Frequencies of words that describe speech and visual imagery during sleep accounted for smaller portions of the REM/NREM variance. To the extent that a simple word frequency recall measure accounts for the same variance that such complex measures as dreaming and imagery share with REM/NREM, we may conclude that judges of Dreaming implicitly rely on a dimension similar to the Total Recall Freq. It is proposed that the cognitive difference between REM and NREM is essentially one of attention and memory. Given the high perceptual thresholds of REM sleep, the events attended to and stored in memory are necessarily unrelated to the surrounding environment and therefore differ from waking thought.

This report summarizes a study of the factorial composition of a series of measures of the structure and content of daydreaming behavior. The relationships of fantasy to measures of divergent thought productivity, attention, curiosity, and personality measures are examined.