Consider the following: An MLB pitcher throws a ball toward you that you need to hit. Presuming that most of our readers count as novices or, at best, less-skilled batters, then ample evidence indicates that – even if the pitcher performed the throw with the exact same movements, and hence would provide the exact same kinematic information as when he faced a professional pinch-hitter – you most likely would not (be able to) attune to and use the same information as the expert (Mann et al. Reference Mann, Williams, Ward and Janelle2007). The events with which you interact, the pitcher's throwing motion and the ball's flight, are identical and thus induce the very same patterns in the optic array, yet the information you attune to for guiding your batting action would be crucially different from the information the expert attunes to and uses. To paraphrase F&S, this is the perfect example for no change in visual input but a dramatic change in visual perception. It underlines our point that perception can be understood only by considering its primary purpose: to guide our actions in and interactions with the environment (Gibson Reference Gibson1979/1986; see also Cañal-Bruland & van der Kamp Reference Cañal-Bruland and van der Kamp2015).

Sticking with the batting example, there is no doubt that you and the professional pinch-hitter differ with respect to movement skills or action capabilities. Consequently, if the ball were to approach you and the expert player at the exact same speed, then the different movement skills would impose differential temporal demands on you with regard to initiating the batting response – that is, you would likely need to initiate the movement earlier than the professional. At the same time, the difference in movement skills, such as the temporal sequencing of the step and the swing, would require that – even though the exact same kinematic patterns were visually available to both – you and the expert player attune to and use different information (e.g., you may have to rely relatively strongly on early movement kinematics, whereas the expert may even be able to exploit early ball flight information).

Put in more general terms, individual action capabilities fundamentally constrain the way actors encounter their environment, and hence, the requirements differ with respect to the information actors need and attend to. In this respect, attention is an actor's attunement to the information that specifies the relevant properties of the environment–actor interaction. This attunement has multiple characteristics: (a) it is task-specific, (b) it is prone to differ within actors across multiple encounters, and (c) it is also prone to differ among actors (e.g., with different levels of skill). These characteristics follow because individuals guide their (inter)actions by accomplishing certain perceptions (Gibson Reference Gibson1979/1986). Let's elaborate on the characteristics of attunement to further specify our argument:

• Attunement is task-specific. Referring back to the batting example: Although when batting, the professional pinch hitter is better attuned than you are to the kinematic information contained in the movements of the pitcher, this baseball player may not be better attuned than you are when he is watching and waiting on-deck before entering as a pinch hitter (Dicks et al. Reference Dicks, Button and Davids2010; Milner & Goodale Reference Milner and Goodale2008; van der Kamp et al. Reference van der Kamp, Rivas, van Doorn and Savelsbergh2008).

• Attunement is prone to differ within actors across multiple encounters. The validity of this argument can be illustrated for different timescales. For example, on their long route to expertise, professional batters converge toward more-specific information available from a pitcher's throwing actions and a ball's flight. That is, with development, learning, or both, those actors who ultimately reach expert levels have learned to attend to information that is simultaneously uniquely specific to and adaptive for their superior batting actions. Within ecological psychology, this process of change is referred to as education of attention (Jacobs & Michaels Reference Jacobs and Michaels2007). Also, on a shorter timescale, such as across multiple throws by the same pitcher within one inning, convergence to more specific information may occur. For example, when a batter faces a new pitcher with a different pitching style, the professional batter may adapt to this style within a few pitches simply by shifting attention to a different invariant within the kinematic information contained in the pitcher's throw – even when the throwing kinematics remain stable over successive pitches.

• Attunement is prone to differ among observers. As indicated above, in various domains such as sports and music, ample evidence demonstrates that given their action capabilities, experts differ as concerns the ability to pick up the most reliable information compared with their less-skilled counterparts (Huys et al. Reference Huys, Cañal-Bruland, Hagemann, Beek, Smeeton and Williams2009; Mann et al. Reference Mann, Abernethy and Farrow2010). For example, expert baseball batters more frequently and effectively use information provided by the pitcher's arm angle and the ball's rotation (Gray Reference Gray2002; Reference Gray2010). Again, these differences in attention across individuals emerge as a result of differences in some combination of adaptation, learning, and development (and perhaps predisposition).

To reiterate, we argue that differences in perception always go together with differences in attention induced by the requirements for adaptive (inter)actions of actors within their environment. This inseparability of perception and action (reflected in attention), in our view, challenges F&S's core claims (1) that perception is an isolable entity or process that functions independently from cognition, attention, and action, and hence (2) that top-down processes do not impinge on perception. F&S's extremely narrow view of what perception is, namely bottom-up processing that results in percepts, neglects the fundamental question what perception is for. Yet, answering the question of what perception is without considering what perception is for naturally – as is the case in F&S's target article – leads to isolated percepts, which may trigger a few contrived issues but are rather marginal, if not meaningless, for understanding situated behaviours.