Psychology of Decision-Making
Psych 🧠 - 363/500
Hello Reader,
Our new initiative, FirstTherapy.org has been launched :)
If you’ve ever thought about going for mental health counselling, this platform offers “Free Mental Health Therapy Sessions”. That too, human-based. Not AI!
What is it?
This research paper investigates how humans make predictions to guide their decisions, specifically exploring the use of
"Forward prediction" (predicting outcomes from a given starting point)
"Backward prediction" (predicting starting points from a given outcome).
The core finding is that humans adaptively employ either forward or backward prediction depending on which is more efficient for the specific task environment.
Major Findings:
Backward prediction is favored in divergent environments: When the number of possible outcomes is much larger than the number of starting states (like a maze with many dead ends and a single solution), backward prediction is more efficient.
In such cases, participants were more likely to start their planning from the desired outcome and work backwards to the starting point.Forward prediction is favored in convergent environments: Conversely, when the number of possible starting states is larger than the number of outcomes (like a maze with multiple paths leading to a single exit), forward prediction is more efficient.
Participants were more likely to plan their actions forward from the initial state to the desired outcome.Humans adaptively deploy both strategies: The research demonstrates that people flexibly switch between forward and backward prediction depending on the task structure, indicating an adaptive strategy to optimize decision-making.
Efficiency over accuracy: Backward prediction can sometimes lead to inaccurate predictions due to its sensitivity to base rates (frequency of different starting states). However, in divergent environments, participants still favored backward prediction, prioritizing efficiency (finding a solution quickly) over accuracy.
Like reading about these insights? Support this Newsletter ❤️
Share The Psych Handbook that has 150+ biases & fallacies explained with emojis!
Or the Amazon Kindle copy from here. (40% off)
What do I need to know:
Humans are flexible prediction machines: We don't just learn what follows a given state; we also learn what precedes a given outcome, adapting our prediction strategy based on which is more efficient.
Environment structure shapes prediction: The type of environment (divergent or convergent) plays a key role in determining whether forward or backward prediction is more useful.
Efficiency can trump accuracy: In complex situations, people may prioritize finding a solution quickly even if it comes at the cost of some inaccuracy.
Implications for decision-making models: The research challenges existing models that primarily focus on forward prediction, suggesting that backward prediction is an important and often overlooked aspect of human decision-making.
Source:
https://osf.io/preprints/psyarxiv/wdbg4