Creating Epistemically Successful Interdisciplinary Research Infrastructures: Translational Cognitive Neuroscience as a Case Study

Jacqueline Anne Sullivan (University of Western Ontario)

Many neuropsychiatric and neurodegenerative diseases and other brain disorders (e.g., concussion) are widely understood to involve impairments in cognition. During the past two decades, neuroscientists have sought to characterize these impairments and identify the neural mechanisms that give rise to them. A common approach has been to investigate the neural mechanisms of cognitive functions in non-human animals using classic paradigms such as fear-conditioning and the Morris water maze and use findings to ground inferences about the causes of cognitive dysfunction in humans. One obstacle to progress in this research area, however, has been the lack of analogous experimental tools to assess cognitive function and dysfunction in experimental animals and humans. Specifically, it is unclear whether experimental tools used in human and non-human animal studies probe the same cognitive functions. 

In this talk, I describe and evaluate a collaborative open-science research initiative, translational cognitive neuroscience (TCN), which began in the 1990s with an eye towards developing a set of complementary experimental tools to investigate cognition in humans and non-human animals. At that time, cognitive neuroscientists were using computer-based touchscreen tasks to assess cognitive impairments in mental illness and neurodegenerative diseases, which were known to correlate with underlying neural dysfunction. With an eye towards facilitating the translation of results from rodent studies to human studies, TCN researchers developed a computer-based experimental apparatus for use with rodents, the rodent touchscreen operant chamber (Bussey et al. 1994), and began developing, refining, and optimizing a set of tests for assessing cognition in rodents that closely resemble touchscreen tests used to assess cognition in humans. 

In this talk, I argue that the success of these investigative tools for producing translational knowledge is contingent on them meeting a number of epistemic benchmarks (e.g., face validity, replicability, construct validity, reliability and robustness). Moreover, ensuring that these benchmarks are met involves an unprecedented amount of coordination within individual laboratories and across research groups. I examine the nature of this coordination and address the question of whether it provides any insights with respect to the correct recipe for cumulative science more generally.