Dement and Kleitman (1957)

Context/ Background

• Pervious research suggested more dream activity in REM than NREM

  • Data was based in self-report

  • Why is this a problem?

  • EEG would allow for objective measure of REM and dreaming

AIM

To find the relationship between REM sleep and dreaming.

Hypotheses

H1- There will be a significant association between REM sleep and dreaming.

H2 - There is a significant positive correlation between the estimate of time spent dreaming and the measuremnet of REM sleep.

H3 - There is a relationship between the pattern of eye movement and reported content of the dream.

Method and Design

• Research Method: “Natural” lab experiment in a controlled setting

• Repreated measures design

• Carried out at the U of Chicago

• IV= Woken in rem or NREM

• DV = Self report of Dreams

• Controls: Standerdized procedure, woken Ps to the same sound of doorbell, used recording device for dream recall

  • Why do we consider a “natural” experiment?

  • The IV’s are naturally occurring

Sample and Participants

• Opportunity Sample

• Participant

  • 7 Males & 2 Females

  • Ps interested to avoid alchohol and caffeine but were told to eat normally

  • Ps reported to the lab just before regular bedtime

  • Slept in a quite dark room.

Data Collection

H1- Quantitative - measured EEG patterns in REM & NREM sleep as reported by Ps recall

H2 - Quantitative - EEG patterns in REM sleep and Ps estimate time of dreaming compared for correlational data (5 vs 15 min)

H3 - Qualitative - Data on dream content compared to observationas on the movement of Ps eyes.

Procedure Testing H1 and H2

• Reposted to lab right before normal

• No alchohol or caffeine; eat normally

• 2-3 electrodes placed nar eyes

• Taken to a quite,dark room to sleep

• EEg run though the night in nextroom

• Woken at various times during REM and NREM with doorbell

• Had to say into recorder in they were dreaming, if it had been 5 or 15 min.

• Only counted if it was detailed and Coherent

• Sometimes Es entered the room ro question them further.

Procedure Testing H3

• The Ps were woken up one minute after 1 to 4 patterns of eye movement

  • A- mainly vertical

  • B - Mainly horizontal

  • C - Both Ver & Hor

  • D - Little / No eye movement

• The Ps were to tell what they were dreaming about

• D&K then analyzed data to see if correlation existed

Results of REM Sleep

• REM never occured at the start of the sleep cycle.

• All Ps showed periods of REM, characterized by a low voltage, relatively fast pattern EEG

• The avg. occurrence of REM sleep was once every as minutes, with range of group between 70-104 min.

• REM lasted between 3-50 min, and increased in time as might progressed.

Results for Hypotheses

• H1: Dream recall occurred mostly in REM vs NREM 

  • Recall for REM 80% vs NREM 7%

• H2: Ps could accurately determine if they had been dreaming for 5 or 15 minutes

  • Correct 83% from 111 awakenings

  • Correlations were found to be significant, ranging from .40 to .71 for each Ps

  • Findings show that we dream relatively in real time

• H3: Findings suggest eye movement is related to dream content 

  • e.g. vertical = climbing a cliff or ladder

  • horizontal = throwing tomatoes

  • both = searching for something or fighting with someone

  • no eye movement = drivin

Conclusion

• Need to be able to say if hypotheses were supported. 

• Please discuss with your partner if you think they were or were not supported.

• Results obtained from waking Ps strongly supports the correlation b/w REM sleep & dreaming

• Based on data, dreams occur in real time

• Dreaming can be objectively measured by recording REM cycles during sleep  

Strenghts and Weaknesses

Strenghts

• Lab exp = more controlled (ex. extraneous variables, no alchohol or caffein, dark room, doorbell, same machine, standerdized procedure)

• No participant varuables due to repeated measures

• Quantitative data

Weaknesses

• Low ecological validity

• Low mundane realism (wires)

• Small sample size = harder to generalize

• Doorbell could have interfered w/ their dreams

• No background on participants

• Correlation does not prove causation

Evaluation: GRAVE

When we evaluate studies we are going to use the GRAVE method.

Generalizability: Problem because of small sample size, nor more males than females.

Reliability: Yes, lots of controls, EEG = objective measurement, quantitative data, standerdized procedure.

Application: Use it to see sleep cycles (are they getting enough), can help with sleep disorders, pilor study helps generate new research.

Validity: Yes, people were abke to say whether they were in REM or NREM, IV was opperationalized, Used EEG to see NREM or REM, DV opperationalized = coherent and detailed

Ethics: No mention of informed consent, Ps to withdraw, no significant harm/stress possible sleep deprivation.