Dream behavior and waking behavior
Do people’s behaviors in dreams correspond to their waking behavior, implicit attitudes, or behavior under low-control conditions?
This idea is a more concrete version of dreams-as-low-control-observation. The core question is not whether particular dream events are common. It is whether a person’s reaction in a dream situation predicts how they react in a meaningfully similar waking situation.
Core challenge
The hardest methodological problem is coding dream behavior in a standardized, reproducible way. Global ratings such as “how violent was this dream?” may be too subjective. Existing dream-content systems are useful, but this project needs special attention to what the dreamer did in the situation.
A promising approach is to code morally or socially relevant events and the dreamer’s role in them:
- aggressor
- victim
- helper
- bystander
- deceiver
- cooperative actor
- avoidant actor
- confrontational actor
The coding system should correct for dream length and number of reports per person.
Design option 1: broad dream-behavior diary
Participants first complete waking measures of aggression, prejudice, or prosociality. They then record dreams for about 30 days using standardized morning instructions.
Measures could include:
- Implicit Association Test for aggressive attitudes or related constructs.
- Buss-Perry Aggression Questionnaire.
- Other-ratings of aggressive or prosocial behavior.
- Socially desirable responding, especially moralistic bias.
- Self-knowledge scales.
- Personal motivation to respond without prejudice.
- Social motivation or external pressure to respond without prejudice.
Dream reports would be coded for relevant behaviors and situations, then correlated with waking measures.
Design option 2: individualized lab scenarios
For stronger ecological validity, researchers could adapt lab scenarios to each participant’s actual dream content. For example, if a participant repeatedly reports conflict with a dream character, a lab task could approximate the same interpersonal structure.
This is expensive and labor-intensive, but it directly targets the correspondence between dream behavior in situation X and waking behavior in situation X.
Lower-budget version:
- Ask for the most recent remembered dream.
- Select participants whose dream situations can be ethically and feasibly approximated in the lab.
- Use a standardized but flexible task such as insult response, resource sharing, helping, or noise-blast aggression.
Higher-budget version:
- Collect 30 days of dream reports.
- Identify frequent, realistic, or easily reproducible situations.
- Build individualized scenarios with help from trained research assistants.
Design option 3: aggression-focused pilot
Start with aggression because it has existing implicit, explicit, other-report, and behavioral measures.
Possible ratios:
- Dream-as-aggressor ratio: dreams where participant is aggressor divided by total dreams.
- Aggressor-to-aggression ratio: participant-as-aggressor events divided by all aggression events.
- Aggression-to-dream ratio: aggression events divided by total dreams.
- Physical-aggression ratio: physical aggression events divided by total dreams or aggression events.
These ratios could be compared with:
- implicit aggression
- explicit aggression
- other-rated aggression
- lab aggression behavior
- Hall and Van de Castle norms where appropriate
Dream collection protocol
Participants should receive standardized instructions to improve recall and detail:
- Set a consistent morning alarm or reminder.
- On waking, pause before doing anything else.
- Write the dream immediately.
- Focus especially on the situation and behavior: “What happened, and what did you do?”
- Record dreams on a simple platform that timestamps entries.
- Optionally use implementation intentions or bedtime self-suggestion to remember dreams.
The protocol should balance detail with burden. Asking for too much detail may reduce compliance or bias reports toward salient dreams.
Compensation logic
Compensation should reflect time burden without encouraging fabrication or excessive padding.
One feasible low-budget model:
- baseline payment for enrollment
- small bonus per dream report
- small bonus per word threshold or completeness threshold
- cap total compensation
For a 30-day diary, expected compliance may be substantially lower than ideal. A planning assumption of roughly 20 dreams per participant with about 200 words each may still be optimistic.
Sample size considerations
Dream content variables can be rare and unstable. The project may need:
- many reports per participant for individual-level estimates
- a large total pool of dreams for content coding
- minimum word-count thresholds for usable reports
- sensitivity analyses for number of dreams, report length, and recall frequency
This should probably begin as a feasibility pilot before a definitive test.
Key references to organize
- Nielsen and Stenstrom (2005) on memory sources of dreaming.
- Zadra and collaborators on dream-content methods and sample-size issues.
- Domhoff on dream content, norms, and sample-size requirements.
- Hall and Van de Castle coding system.
- Work on made-up dream detection and dream recall motivation.
- Work on implicit aggression measures and their relation to self-report and behavior.
Key risks
- Dream reports may reflect recall motivation, salience, or narrative style more than behavior.
- Incentives could increase fabricated or embellished reports.
- Coding dream behavior requires strong inter-rater reliability and clear decision rules.
- Individualized lab scenarios increase ecological validity but reduce standardization.
- Dream content should not be moralized or treated as direct evidence of character.
Related ideas: