I am a philosopher of neuroscience from Pittsburgh, Pennsylvania (I am thus of course a big fan of the Pens and the Steelers). I majored in philosophy as an undergraduate at Clark University in Worcester, Massachusetts, where I wrote an honours thesis on Southern novelist Walker Percy’s treatment of mental illness and the self. I did my PhD work in the Department of History and Philosophy of Science (HPS) at the University of Pittsburgh (Pitt). Although I thought I would pursue work in philosophy of psychiatry at Pitt, I became interested in neuroscience and decided to pursue a MS in neuroscience in the Department of Neuroscience at Pitt. There, I worked with Edda (Floh) Thiels, who trained me in in vivo electrophysiological techniques and Bea Kanterewicz, who trained me in Western blot analysis. For my 2.5 year MS project, I investigated the role of the extracellular signal-regulated kinase (ERK) in long-term potentiation (LTP) in area CA1 of the rat hippocampus in vivo. Long-term potentiation historically has been thought to be the Hebbian-like (“cells that fire together wire together”) synaptic mechanism that underlies learning and memory. The aim of my MS project was to compare ERK activation in LTP to its activation in long-term depression (LTD) in order to determine if ERK was a general or specific plasticity kinase. My MS work was funded through a National Science Foundation Integrative Graduate Education and Research Training Grant (NSF IGERT) awarded to me by the Center for the Neural Basis of Cognition (Pitt-CMU). I was also the first PhD student in HPS permitted to enroll in the CNBC’s Graduate Training Program and the first ever to earn a CNBC Graduate Training Certificate.

The in vivo electrophysiological and biochemical experiments that I ran as a MS student were challenging; each experiment involved a number of different steps and at each step confounds or errors that might compromise the reliability of the data production process were possible. This prompted my interest in reliability as an epistemic constraint on the experimental process. I also was interested in the question of what my experiments actually indicated about the mechanisms of LTP as it was being investigated in other laboratories (external validity)–this was in part because different labs used different LTP-inducing stimulation protocols and it was known that such differences resulted in differences in the cellular and molecular mechanisms by which LTP was produced (e.g., See Dudek & Fields 2001) and thus it was an open question whether these different labs were looking at the same phenomenon or different phenomena. I was also interested in what understanding the mechanisms of LTP and LTD actually indicated about the mechanisms of learning and memory phenomena (external validity/ecological validity). I addressed these issues in my PhD dissertation and published the core ideas from my dissertation in a publication in Synthese that was published online in 2008. These questions were novel in part because philosophers of neuroscience at that time (e.g., Bickle 2006; Craver 2007) were interested in LTP and memory as case studies for thinking about reduction and mechanistic explanation, but as I argued in my paper, these same philosophers had paid scant attention to whether and how LTP and memory experiments produce knowledge.






My work since 2009 has continued to have a focus on experimental practice in neuroscience, but it has branched out in a number of different directions. For example, one of the issues that I addressed in my dissertation and have since published on is the issue of construct validity (e.g., Cronbach & Meehl 1955; Shadish, Cook & Campbell 2001)– whether experimental apparatuses, tasks or paradigms in neuroscience actually measure what cognitive capacity they are intended to measure and what the implications are when they don’t and they are imported into areas of science that are aimed at illuminating mechanisms and not interested in understanding ‘what’ cognitive capacity is being detected and measured. I have a paper on the Morris Water Maze (Sullivan Synthese 2010) in which I investigate this question. In two other papers, I argue that the concept of representation ought to play a role in determining “what” an organism learns when it is trained in a paradigm like the hidden condition of the Morris Water Maze and in a subsequent paper (Sullivan 2014, contribution to a great anthology edited by Charles Wolfe, Brain Theory (Palgrave-Macmillan 2014)) that was based on a collaborative project I was doing with Floh Thiels (Sullivan and Thiels 2011, SFN poster presentation), I argue that it is important for rodent behavioral researchers to take “the intentional stance” (e.g., Dennett 1987) in the context of designing and implementing their experiments in order to determine “what” rodent subjects trained in experimental paradigms and tasks are actually learning. A related paper (Sullivan 2016) contains the figure below, and emphasizes the importance of construct validation and explication (Shadish, Cook & Campbell) for progress in the neurosciences of cognition. This paper was a contribution to an awesome anthology on Natural Kinds and Classification in Science (Routledge 2016) edited by the excellent Philosopher of Science Catherine Kendig https://philosophy.msu.edu/faculty-staff/catherine-kendig/.






PDFs of all of my papers are available on PhilPapers. And here is a nice write-up of my earlier work in the SEP entry for Philosophy of Neuroscience (2018) written by Valerie Hardcastle and John Bickle.