Thread by @PeterKvam, Cool, that makes things a little easier. Here's a quick, incomplete list, [...]

Cool, that makes things a little easier. Here& #39;s a quick, incomplete list, including some that are actually incorrect predictions.
Example 1: A choice can only be triggered after sampling a piece of evidence for the to-be-chosen option [sorta incorrect]
https://www.sciencedirect.com/science/article/pii/0022249688900429">https://www.sciencedirect.com/science/a... https://twitter.com/DobyRahnev/status/1379130150722228226">https://twitter.com/DobyRahne...

Example 2: Choice is determined by the balance of evidence, not support for any individual option, so the magnitude of support / total evidence for different options should not affect choice as long as balance is maintained [incorrect] https://www.nature.com/articles/s41598-019-56392-0">https://www.nature.com/articles/...

Sensitivity of reaction time to the magnitude of rewards reveals the cost-structure of time

https://www.nature.com/articles/s41598-019-56392-0

Example 3: The distribution of evidence sampled by decision makers following a diffusion strategy should not match the true distribution of evidence - it should be more extreme [correct, and my personal recent favorite] https://psyarxiv.com/h9zcv/ ">https://psyarxiv.com/h9zcv/&qu...

Polarization and extremism emerge from rational choice: Estimation as a solution to biased sampling

Polarization is often described as the product of biased information search, motivated reasoning, or other psychological biases. However, polarization and extremism can still occur in the absence of...

https://psyarxiv.com/h9zcv/

Example 4: Because it is based on a classical random walk, measuring the system (i.e., eliciting a choice / judgment) should not change its state, [incorrect, based on quantum cognition work]
https://www.pnas.org/content/112/34/10645.short
https://www.pnas.org/content/1... href=" https://www.nature.com/articles/s41598-019-54383-9">https://www.nature.com/articles/... https://psyarxiv.com/cb26p ">https://psyarxiv.com/cb26p&quo...

Temporal oscillations in preference strength provide evidence for an open system model of constru...

The decision process is often conceptualized as a constructive process in which a decision maker accumulates information to form preferences about the choice options and ultimately make a response....

https://www.pnas.org/content/112/34/10645.short

Example 5: This isn& #39;t totally explicit, but if you buy into the log-odds interpretation of the evidence state (some don& #39;t), then drift should be equally responsive to manipulations of sample size and sample proportion [incorrect]
https://www.sciencedirect.com/science/article/pii/001002859290013R">https://www.sciencedirect.com/science/a... https://www.sciencedirect.com/science/article/pii/S0010027716300968">https://www.sciencedirect.com/science/a...

Strength and weight: The determinants of choice and confidence

Evidence for different hypotheses is often treated as a singular construct, but it can be dissociated into two parts: its strength, the proportion of …

https://www.sciencedirect.com/science/article/pii/001002859290013R

Example 6: Evidence represented by decision makers should not be autocorrelated across time if the underlying information itself isn& #39;t autocorrelated (Markov property) [incorrect - Rachel Heath has been working on this a lot] https://journals.sagepub.com/doi/abs/10.1080/713755987">https://journals.sagepub.com/doi/abs/1...

Response-Time Dynamics: Evidence for Linear and Low-Dimensional Nonlinear Structure in Human Choice...

Response time (RT) is a commonly used measure of cognitive performance, which is usually characterized as stochastic. However, useful information may be hidden ...

https://journals.sagepub.com/doi/abs/10.1080/713755987

Example 7: Absorbing choice boundaries imply that once a choice is made (where having made a choice has a monotonically increasing probability over time), subsequent evidence is ignored. [mostly correct - there& #39;s a cool recent neurofeedback paper on this] https://doi.org/10.1038/s41586-020-03181-9">https://doi.org/10.1038/s...

Decoding and perturbing decision states in real time

In macaque motor cortex, moment-to-moment fluctuations in neurally derived decision variables are tightly linked to decision state and predict behavioural choices with better accuracy than conditio...

https://doi.org/10.1038/s41586-020-03181-9

Of course, modifications like collapsing bounds can make it work in some scenarios, but I& #39;ve been surprised by how well the DDM has stuck around despite failed predictions (truthfully, I still use it..). Maybe part of the problem is that the alternatives are hard to work with?

Latest Threads Unrolled: