Firestone & Scholl (F&S) argue that the evidence that allegedly shows that cognition affects visual perception, once properly examined, does not support the view that the percept, the outcome of vision, is directly affected in a top-down way by cognition. By “direct cognitive effects,” I mean, extending the authors' view (sect 4.5.1, para. 1), those cognitive influences that affect perceptual processing itself and change the percept, and not the cognitive effects that determine to what or where the perceiver attends. The authors state (sect. 1.2, para. 2) that whereas many previous discussions defended the modular nature of only a circumscribed, possibly unconscious, stage of visual processing – that is, “early vision” – they aim to assess the evidence for top-down effects on perception as a whole, including the conscious percept. Therefore, their discussion encompasses the cognitive effects on both early and late vision, the latter being the stage in which the percept is constructed.
Considering perception as a whole, the authors do not distinguish between cognitive effects on early vision and cognitive effects on late vision. This poses a problem because cognition affects early vision and late vision differently. I agree that early vision is cognitively impenetrable because there are no direct cognitive effects on early vision. There is, however, substantial neuropsychological evidence that late vision and, hence, the percept is directly cognitively penetrated because the perceptual processes of late vision use cognitive information as a resource and, thus, cognition modulates the processes of late vision.
The authors do not appreciate that late vision is cognitively penetrated because they restrict themselves to considering, among the various forms of attention, only peripheral attention – that is, attention as a determinant of the locus or object/feature of focus. The authors acknowledge that attentional effects that are not peripheral exist, and that attention may “interact in rich and nuanced ways with unconscious visual representations to effectively mold and choose a ‘winning’ percept – changing the content of perception rather than merely influencing what we focus on” (sect. 4.5, para. 6), but they opt to focus on peripheral attention.
Peripheral attention selects the input to perception but does not affect how the processing operates (ibid, sect. 4.5.1, para. 1). Accordingly, the authors correctly conclude that when perceptual behavior is explained by invoking the role of peripheral attention, even if peripheral attention is guided by cognitive states, that does not entail perception being cognitively penetrated, a view also shared by the majority of philosophers (Zeimbekis & Raftopoulos Reference Zeimbekis and Raftopoulos2015). Attention, however, especially if viewed in line with the bias competition model, does not act only in this external way. Rather than merely selecting input, attention is integrated with, and distributed across, visual processing (Mole Reference Mole, Ziembekis and Raftopolous2015; Raftopoulos Reference Raftopoulos2009). Attentional effects are intrinsic in late vision rendering it cognitively penetrated (Raftopoulos Reference Raftopoulos2009; Reference Raftopoulos2011).
There are several ways cognition affects late vision, such as the application of concepts on some output of early vision so that hypotheses concerning the identities of distal objects can be formed and tested in order for the objects to be categorized and identified. Cognitively driven attention is one way – an often-studied way – cognition affects perceptions. Research suggests that cognitively driven spatial and object/feature-centered attention, as expressed by the N2 Event Related Potential (ERP) component, affects perceptual processing. The N2 is elicited about 200–300 ms after stimulus onset in monkeys and humans, in the area V4 and in the inferotemporal cortex. Research (Chelazzi et al. Reference Chelazzi, Miller, Duncan and Desimone1993; Luck Reference Luck1995) suggests that the N2 reflects the allocation of attention to a location or object and is influenced by the type of the target and the density of the distractors. It is also sensitive to stimulus classification and evaluation (Mangun & Hilyard Reference Mangun, Hilyard, Rugg and Coles1995). Thus, N2 is considered to be a component of cognitively driven or sustained attention.
Cognitively driven attention, for example, affects color processing in the human collateral sulcus (the main cortical area for analysis and coding of color information) at about 160 ms (Anllo-Vento et al. Reference Anllo-Vento, Luck and Hillyard1998). At about 235 ms in primate V1, attention distinguishes target from distractor curves (Roelfsema et al. Reference Roelfsema, Lamme and Spekreijse1998). Attention is thought to enhance the activity of neurons in the cortical regions that encode the attended stimuli. The timing of these cognitive effects places them within late vision but outside early vision, which means that late vision, but not early vision, is affected directly by cognition. It should be noted that the various precueing effects that affect early vision processing are not direct but indirect cognitive influences because they do not affect perceptual processing itself but, rather, the preparatory neuronal activity of the perceptual circuits (Raftopoulos Reference Raftopoulos2015).
Why does attention enhance the activity of some neurons in the visual cortical regions during late vision? Clark (Reference Clark2013) argues that to perceive the world is to use what you know to explain away the sensory signal across multiple spatial and temporal scales; the process of perception is inseparable from cognitive processes. The aim of this interplay is to enable perceivers to respond and eventually adapt their responses as they interact with the environment so that this interaction be successful. Success in such an endeavor relies on inferring correctly (or nearly so) the nature of the source of the incoming signal from the signal itself.
Current research sheds light on the role of top-down cognitive effects in inferring correctly the identities of the distal objects during late vision. The cognitively driven direct attentional effects within late vision contribute to testing hypotheses concerning the putative distal causes of the sensory data encoded in the lower neuronal assemblies in the visual processing hierarchy. This testing assumes the form of matching predictions, made on the basis of a hypothesis, about the sensory information that the lower levels should encode assuming that the hypothesis is correct, with the current, actual sensory information encoded at the lower levels (Barr Reference Barr2009; Kihara & Takeda Reference Kihara and Takeda2010; Kosslyn Reference Kosslyn1994). To this aim, attention enhances the activity of neurons in the cortical regions that encode the stimuli that most likely contain information relevant to the testing of the hypothesis.
Firestone & Scholl (F&S) argue that the evidence that allegedly shows that cognition affects visual perception, once properly examined, does not support the view that the percept, the outcome of vision, is directly affected in a top-down way by cognition. By “direct cognitive effects,” I mean, extending the authors' view (sect 4.5.1, para. 1), those cognitive influences that affect perceptual processing itself and change the percept, and not the cognitive effects that determine to what or where the perceiver attends. The authors state (sect. 1.2, para. 2) that whereas many previous discussions defended the modular nature of only a circumscribed, possibly unconscious, stage of visual processing – that is, “early vision” – they aim to assess the evidence for top-down effects on perception as a whole, including the conscious percept. Therefore, their discussion encompasses the cognitive effects on both early and late vision, the latter being the stage in which the percept is constructed.
Considering perception as a whole, the authors do not distinguish between cognitive effects on early vision and cognitive effects on late vision. This poses a problem because cognition affects early vision and late vision differently. I agree that early vision is cognitively impenetrable because there are no direct cognitive effects on early vision. There is, however, substantial neuropsychological evidence that late vision and, hence, the percept is directly cognitively penetrated because the perceptual processes of late vision use cognitive information as a resource and, thus, cognition modulates the processes of late vision.
The authors do not appreciate that late vision is cognitively penetrated because they restrict themselves to considering, among the various forms of attention, only peripheral attention – that is, attention as a determinant of the locus or object/feature of focus. The authors acknowledge that attentional effects that are not peripheral exist, and that attention may “interact in rich and nuanced ways with unconscious visual representations to effectively mold and choose a ‘winning’ percept – changing the content of perception rather than merely influencing what we focus on” (sect. 4.5, para. 6), but they opt to focus on peripheral attention.
Peripheral attention selects the input to perception but does not affect how the processing operates (ibid, sect. 4.5.1, para. 1). Accordingly, the authors correctly conclude that when perceptual behavior is explained by invoking the role of peripheral attention, even if peripheral attention is guided by cognitive states, that does not entail perception being cognitively penetrated, a view also shared by the majority of philosophers (Zeimbekis & Raftopoulos Reference Zeimbekis and Raftopoulos2015). Attention, however, especially if viewed in line with the bias competition model, does not act only in this external way. Rather than merely selecting input, attention is integrated with, and distributed across, visual processing (Mole Reference Mole, Ziembekis and Raftopolous2015; Raftopoulos Reference Raftopoulos2009). Attentional effects are intrinsic in late vision rendering it cognitively penetrated (Raftopoulos Reference Raftopoulos2009; Reference Raftopoulos2011).
There are several ways cognition affects late vision, such as the application of concepts on some output of early vision so that hypotheses concerning the identities of distal objects can be formed and tested in order for the objects to be categorized and identified. Cognitively driven attention is one way – an often-studied way – cognition affects perceptions. Research suggests that cognitively driven spatial and object/feature-centered attention, as expressed by the N2 Event Related Potential (ERP) component, affects perceptual processing. The N2 is elicited about 200–300 ms after stimulus onset in monkeys and humans, in the area V4 and in the inferotemporal cortex. Research (Chelazzi et al. Reference Chelazzi, Miller, Duncan and Desimone1993; Luck Reference Luck1995) suggests that the N2 reflects the allocation of attention to a location or object and is influenced by the type of the target and the density of the distractors. It is also sensitive to stimulus classification and evaluation (Mangun & Hilyard Reference Mangun, Hilyard, Rugg and Coles1995). Thus, N2 is considered to be a component of cognitively driven or sustained attention.
Cognitively driven attention, for example, affects color processing in the human collateral sulcus (the main cortical area for analysis and coding of color information) at about 160 ms (Anllo-Vento et al. Reference Anllo-Vento, Luck and Hillyard1998). At about 235 ms in primate V1, attention distinguishes target from distractor curves (Roelfsema et al. Reference Roelfsema, Lamme and Spekreijse1998). Attention is thought to enhance the activity of neurons in the cortical regions that encode the attended stimuli. The timing of these cognitive effects places them within late vision but outside early vision, which means that late vision, but not early vision, is affected directly by cognition. It should be noted that the various precueing effects that affect early vision processing are not direct but indirect cognitive influences because they do not affect perceptual processing itself but, rather, the preparatory neuronal activity of the perceptual circuits (Raftopoulos Reference Raftopoulos2015).
Why does attention enhance the activity of some neurons in the visual cortical regions during late vision? Clark (Reference Clark2013) argues that to perceive the world is to use what you know to explain away the sensory signal across multiple spatial and temporal scales; the process of perception is inseparable from cognitive processes. The aim of this interplay is to enable perceivers to respond and eventually adapt their responses as they interact with the environment so that this interaction be successful. Success in such an endeavor relies on inferring correctly (or nearly so) the nature of the source of the incoming signal from the signal itself.
Current research sheds light on the role of top-down cognitive effects in inferring correctly the identities of the distal objects during late vision. The cognitively driven direct attentional effects within late vision contribute to testing hypotheses concerning the putative distal causes of the sensory data encoded in the lower neuronal assemblies in the visual processing hierarchy. This testing assumes the form of matching predictions, made on the basis of a hypothesis, about the sensory information that the lower levels should encode assuming that the hypothesis is correct, with the current, actual sensory information encoded at the lower levels (Barr Reference Barr2009; Kihara & Takeda Reference Kihara and Takeda2010; Kosslyn Reference Kosslyn1994). To this aim, attention enhances the activity of neurons in the cortical regions that encode the stimuli that most likely contain information relevant to the testing of the hypothesis.