Skip to main content

Similarity in the making: how folk psychological concepts facilitate development of psychological concepts


This paper draws on the notion of “objects of research” in psychology as clusters of phenomena (Feest in Philos Sci 84:1165–1176, 2017) to analyze the productive role of folk psychological concepts—and the operational definitions that arise from them—in the development of concepts in scientific psychology. Using the case study of similarity, I discuss the role of the folk psychological concept in the regimentation of different measures of similarity judgments. I propose that by giving rise to operational definitions that lead to experimental dissociation on the one hand, and by providing the concept with unity on the other hand, the folk psychological concept generates a productive tension that facilitates empirical and theoretical development of the scientific concept.

This is a preview of subscription content, access via your institution.


  1. In this paper, I use the term “operational definitions” to refer to “paradigmatic conditions of application” for a concept, which allow scientists to individuate experimentally their objects of research in order to launch their investigation (Feest 2010, 178; see also Sullivan 2009, Bloch 2012). The use of the term “definition” does not imply that such definitions convey the concept’s meaning (Chang 2021, 2017; Feest 2010).

  2. Sullivan (2016) distinguishes between production procedures, which specify the stimuli, measurement procedures, which specify the variables to be measured, and detection procedures, which provide criteria for the measures to be interpreted as indicators of the cognitive capacity under investigation. In what follows, I use “measures” or “measurement procedures” of similarity judgments more broadly, but my focus will be on what Sullivan classifies as measurement and detection procedures.

  3. Feest’s (2017) usage of “phenomenon” is, therefore, more restricted than that of Bogen and Woodward (1988), allowing her to distinguish between repeatable regularities on the one hand and a stable kind with repeatable characteristics on the other hand.

  4. In the context of psychological research, I use “similarity” and “similarity judgments” interchangeably. Both refer to the grasp of a relation by an observer and not to any observer-independent relation.

  5. See also Sullivan (2009) for a discussion of the way in which operational definitions are built into the experimental protocol.

  6. As Feest, (unpublished) points out, this is in line with Carnap’s account of explication. Carnap writes that explication is “the transformation of an inexact, prescientific concept, the explicandum, into an exact concept, the explicatum. The explicatum must fulfil the requirements of similarity to the explicandum, exactness, fruitfulness, and simplicity” (Carnap 1962, 1). Accordingly, while the new concept is more exact and fruitful, its explanatory value with respect to the older concept depends on its continuity with that concept.

  7. These comments are made in the context of Tversky’s discussion of the directionality of similarity judgments, a topic which I will not go into here. The present point is that he appeals to similarity statements in natural language in order to investigate the nature of similarity.

  8. For further discussion of this point, see Medin and Ortony (1989) and Medin et al. 1993.

  9. These procedures were used broadly by scientists with varying additional theoretical commitments. For example, similarity ratings were used both by psychologists who thought that similarity is calculated on the basis of discrete common and distinctive features (e.g., Tversky 1977) and by those who thought it is calculated over the distance between objects along common dimensions in psychological space (e.g., Torgerson 1965).

  10. For an example of similarity considered as a singular outcome, see, again, the quote from Sloman and Rips (1998) in the previous section. It should be noted that even approaches that take similarity as context- and task- dependent often treat it as an output in the sense discussed above, even as they incorporate contextual cues into the factors that affect the output (e.g., Tversky 1977).

  11. As Goldstone and Medin (1994) acknowledge, however, a process model of similarity alone cannot account for all the differences between the tasks associated with the various measures of similarity, as these tap into different processes with nontrivial differences.

  12. See also Goldstone and Medin (1994).

  13. Even within the unifying model, individual measures are associated with specific phenomena that involve varying degrees of processing and that are affected by specific task-related constraints (see discussion in Goldstone and Medin 1994).


Download references


I would like to thank Uljana Feest and Anthony Peressini, as well as two anonymous reviewers, for helpful comments on previous versions of this manuscript.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Corinne L. Bloch-Mullins.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Bloch-Mullins, C.L. Similarity in the making: how folk psychological concepts facilitate development of psychological concepts. Synthese 200, 76 (2022).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI:


  • Scientific concepts
  • Conceptual development
  • Objects of research
  • Folk psychology