Iliana Samara
Assistant Professor · Cognitive Psychology
Cognitive Psychology Unit, Leiden University
I study how people infer interest, attraction, and intention under uncertainty, combining behavioral experiments with signal detection theory, multilevel modeling, and computational approaches.
Cognitive scientist of attraction, social inference & decision under uncertainty.
The problem I study
People rarely know what others feel, want, or intend, but they nonetheless act on those guesses.
A small demonstration
Someone smiles, holds eye contact a little longer than expected, and keeps the conversation going.
The point is not that one answer is correct. The same cue supports different judgments depending on context, prior expectations, and the threshold someone uses before acting. Read that line loosely for everyone and you get what the literature calls the sexual overperception bias; read it strictly and you miss real interest. Where people place that line is part of what I study. The harder problem, and the argument of my paper in Evolution and Human Behavior, is that the line by itself does not tell you why it sits where it does. Curious how far this goes? Play with the interactive demos, including a two-minute game that measures your own criterion.
Research
My research asks how attraction shapes attention, interpretation, and social decisions when the signal is noisy and the stakes are personal. I work across sexual overperception, consent clarity, mimicry and synchrony, pair-bonding, and comparative social cognition. A theme running through all of it is identifiability: the same observed judgment can be produced by several different mechanisms, so studies have to be designed to tell them apart.
Recent
-
2026
New paper
When friendliness looks like desire is out in Nature Reviews Psychology, revisiting Haselton (2003) and what it takes to show that errors in reading sexual interest are directional.
-
2026
New paper
Sexual overperception as a proximate identification problem is out in Evolution and Human Behavior, open access. The same overperception pattern can come from sensitivity, prior beliefs, payoffs, or reporting policy, and an audit of 54 studies found that most designs cannot tell those routes apart.
-
2026
New preprint
When felt states become social evidence sets out an identifiability framework for hot-state social inference. Preprint on PsyArXiv.
-
2026
New paper
Predict this: minimal commitments for testable predictive processing explanations is out in Neuroscience & Biobehavioral Reviews.
-
2026
New on the site
Interactive demos of the ideas behind my research: measure your own d′ and criterion in a two-minute game, then explore signal detection, error management, and Bayesian updating in the browser.
-
2026
New paper
A practical tutorial on signal detection theory with multilevel models for dyadic judgments is out in Methods in Psychology, with an interactive app to go with it.
-
2026
New paper
Estimating consent clarity requires sampling absence, refusal, and withdrawal appears in Archives of Sexual Behavior.
-
2026
New project
Helping build a new consortium and research network on cybersecurity at the faculty.