The target article by Murayama and Jach (M&J) does exactly what one wants such a piece to do – it makes one step back and rethink the broad assumptions and premises that guide one's work. Such a meta-level piece, especially one so thoughtfully and even provocatively articulated, can be extremely helpful in clarifying one's perspective and laying out guidelines and priorities for how to proceed. The piece fits in a long and admirable tradition of internal criticism of the way motivation constructs are conceptualized and utilized to explain behavior (for other noteworthy examples, see Bindra, Reference Bindra1959; Bolles, Reference Bolles1978; Brown, Reference Brown1961; Cofer, Reference Cofer1972; Kantor, Reference Kantor1942; Kleinginna & Kleinginna, Reference Kleinginna and Kleinginna1981). Although this form of critique is not new per se, we believe it is important and needed at present in motivation science.

The authors' critique centers on high-level motivation constructs, and two core premises of the critique are that (1) high-level motivation constructs are not clearly defined and conceptualized, and (2) the nature of the influence of high-level motivation constructs on behavior is not well-understood. That is, “what they are” and “how they work” are not clearly specified (p. 25). Regarding the first premise, we fully agree and simply offer elaboration. Good conceptualization, in the motivation literature and beyond, requires clear construct definition and clear articulation of the construct's functional role. Slippage on the conceptual (as well as operational) front produces what we sometimes see in the motivation literature – jingle-jangle fallacies and resultant inconsistent empirical literatures that are difficult to summarize and interpret. We think clarity here begins with drawing an explicit and precise distinction between the two basic components of motivation – energization and direction. We define energization as the initial instigation of behavior that orients in a general way (the “why”; e.g., needs/motives) and direction as the channeling of this energization toward a specific end state (the “how”; e.g., goals/tactics; Elliot, Reference Elliot, Bong, Reeve and Kim2023). Critically, once separated and carefully conceptualized, energization and direction must be put back together for a full and complete motivational explanation of behavior. For example, neither the need for competence (energization) nor achievement goals (direction) are sufficient to account for achievement behavior; both are needed in combination. We have called this combined construct a “goal complex” (Elliot & Thrash, Reference Elliot and Thrash2001) – the integrated representation of a focal goal (direction) and the reason behind the goal that prompted its adoption (energization) – and a growing body of research attests to the theoretical and empirical utility of this concept (for reviews, see Liem & Senko, Reference Liem and Senko2023; Sommet, Elliot, & Sheldon, Reference Sommet, Elliot, Sheldon, Robbins and John2021).

Regarding the second premise, M&J call for a focus on lower-level processes in motivational analyses of behavior, arguing that constructs at this level are optimally suited to explain behavior. Here we both agree and disagree. We agree that comprehensive motivational explanations must include lower-level processes. Indeed, we view motivation as decidedly hierarchical (Elliot, Reference Elliot2006; see also Cacioppo & Berntson, Reference Cacioppo and Berntson1994; Carver & Scheier, Reference Carver and Scheier2001; Gallistel, Reference Gallistel1982), encompassing myriad constructs at many levels across the neuraxis. In the main, the authors seem to argue that lower-level processes are understudied and such processes are needed to complement the explanatory value provided by high-level constructs (e.g., “No level of understanding should be dismissed as ‘wrong’ [i.e. one level of explanation should not be replaced with a lower-level explanation], because they just explain the behavior for different purposes,” p. 21); we fully agree. At other points, however, they seem to argue for lower-level processes as a replacement for high-level constructs (e.g., “It is the mental computational processes, not the motivation constructs themselves, that are necessary to understand human behavior,” p. 15); here we disagree. We see the value of explicating lower-level processes, but not at the expense of high-level constructs. Many levels of explanation needed – from rudimentary exteroceptive reflexes to subcortical computations to cortical appraisals to the emergent high-level constructs that the authors critique. We think all of these levels of analysis are worthy of study and provide added value to motivational explanations of behavior. Selecting the lower-level as the key to explanation seems to run the risk of reductionism (Sheldon, Reference Sheldon2004), if not infinite regress (Boden, Reference Boden1972). This level of analysis issue is reminiscent of Tolman's (Reference Tolman1932) critique of behaviorism's sole focus on the “molecular” and his advocating for an additional, “molar,” level of analysis that incorporates purpose; he described this molar level as “emergent” and as being “more than and different from the sum of” its molecular parts (p. 7). Categorizing high-level constructs as emergent or even psychologically constructed (as M&J do) does not necessarily mean they are epiphenomenal (as they seem to imply). Such high-level constructs can have an important, independent influence on behavior, often via evocation or recruitment of lower-level processes (e.g., for empirical evidence regarding the aforementioned goal complexes, see Sommet & Elliot, Reference Sommet and Elliot2017). In short, each level of analysis has value, and the optimal level at any given time depends on one's overarching aim (e.g., to acquire a deeper understanding of underlying processes, to discover when and how to intervene, to explain a phenomenon to laypeople, etc.).

Psychological constructs are scientific tools used to describe, categorize, and organize collections of observations, and theories represent integrated combinations of such constructs. Human beings and the behavior they emit are extraordinarily intricate and complex; good theories must, by necessity, match this intricacy and complexity (i.e., be “level adequate,” see Berridge, Reference Berridge2004, p. 17). In motivation science, we believe the best theories will be those that are unabashedly hierarchical, thoroughly inclusive, and deeply integrative (Elliot & Sommet, Reference Elliot and Sommet2023), comprised of both well-defined high-level motivation constructs and multiple levels of lower-level psychological processes. We concur with M&J that a major focus of research in motivation science moving forward needs to be on lower-level processes (of many sorts at many levels). We hasten to add that this work will only advance motivation science to the degree that the findings from it are carefully and thoughtfully integrated into the existing work on high-level constructs.