Statistical cognition
The notion that learning is a matter of developing and testing hypotheses has been applied successfully to several spheres of human development, including social cognition and intertemporal choice. Formal quantitative frameworks based on this notion treat learning as process of statistical inference. Evidence from temporal learning experiments demonstrates that similar frameworks can predict changes in the behaviour of other animals. Experiments evaluating hypothesis testing as the engine driving rats' and pigeons’ temporal learning will reveal the mechanisms driving maladaptive statistical cognition in humans, from impulsive choice to problem gambling and beyond.
Power analysis and experimental design
Statistical power is an important consideration in empirical research: how much evidence do I need to ensure that the conclusions about my results, whatever they may be, are convincing? Practical guidelines and user-friendly software make it possible for behavioral researchers to incorporate statistical power into certain types of small-N group designs (Kyonka, 2019). A priori power analysis involves estimating the sample size required for a study based on predetermined maximum tolerable Type I and II error rates and the minimum effect size that would be clinically, practically, or theoretically meaningful.
Similar issues exist in single-subject research. How many observations are necessary to verify a stable baseline or evaluate whether an intervention has affected behavior? How long should sessions last? Behavior analysts often rely on historical precedent from extant literature or lab lore to make such decisions, but should we?
Similar issues exist in single-subject research. How many observations are necessary to verify a stable baseline or evaluate whether an intervention has affected behavior? How long should sessions last? Behavior analysts often rely on historical precedent from extant literature or lab lore to make such decisions, but should we?