Ito, Maldonado et al. (2022) ascertain (and we believe for all mammals once consulting the literature) that in order to study even area V1 (classically a sensory area, Schiller and Tehovnik 2015), one must allowed the subject (e.g., a mouse, a cat, a monkey, or a human) to move freely to enable volitional behaviors to be expressed to best map the receptive fields of the neurons This notion goes back to the groundbreaking study of Held and Hein (1963) who demonstrated, beyond a doubt, that if kittens are given the chance to move around to explore their visual world, then (and only then) do they learn to perform depth perception, but in the absence of this exploration (by having a kitten’s movements restricted) kittens fail to develop a capacity for depth perception. As we now know, this learning must occur during a critical period shortly after birth, otherwise stereovision does not develop in mammals (Hubel and Wiesel 1977).

In the paper of Ito, Maldonado et al. (2022) it is found that as compared to when a monkey is required to fixate a spot of light while visual images are being presented, if instead a monkey is allowed to free-view the image, the responsivity of V1 cells to visual stimuli is enhanced, especially for neurons situated in laminae II and III, which are the layers that are affected by feedback from the frontal and other lobes in mammals (Froudarakis et al. 2019; Lamme 1995; Lamme and Roelfsema 2000; Leinweber et al. 2017; Schnabel et al. 2018; Zhang et al. 2014; Zipser et al. 1996). At least in the mouse, the anteromedial cortex, which mediates body, head, and eye movements (Tehovnik, Patel, Tolias et al. 2021), sends a direct projection to V1 via the cingulum that modulates the activity of V1 neurons (Niell and Stryker 2010; Zhang et al. 2014). Also, this modulation begins some 125 ms after the onset of a visual stimulus and continues for the duration of the visual analysis out to 400 ms and beyond if needed (Lemme 1995; Zipser et al. 1996). The enhancement of activity for V1 neurons in the Ito, Maldonado et al. study occurs following a ‘spontaneous’ fixation period of 125 ms or more (see Fig. 2B, monkeys 1 & 2 of Ito, Maldonado et al. 2023). This provides sufficient time for the feedback signal to be transmitted from the frontal lobes (as well as from the temporal lobes, Schiller and Tehovnik 2015) to area V1 for the ‘conscious’ identification of a stimulus. For latencies shorter than 125 ms (which are express latencies characteristic of express saccades) there would be no time for the frontal lobes to participate in the neural response and hence the cognitive identification of the stimulus. In this case, we suspect that the signal after V1 bypasses all frontal/temporal lobe participation and is transmitted via the superior colliculus directly to the cerebellum/brain stem to evoke a saccade, after which there might be or might not be a visual analysis by neocortical association areas, which depends on the learning history of an animal and what it has stored away in its neocortex, declaratively. For example, presenting an unfamiliar object during an express fixation of an object (i.e., a fixation of less than 125 ms) should fail to be identified ‘consciously’ by a subject whether a mouse, a cat, a monkey, or a human; on the other hand, the identification of a familiar object will only occur using ‘subconscious’ pathways during an express fixation, which are pathways at and beyond and including the superior colliculus/pretectum and the cerebellum (De Haan, Lamme et al. 2020; Tehovnik, Hasanbegović, Chen 2024; Tehovnik, Patel, Tolias et al. 2021).

We know that the shortest signal transmission from V1 of monkeys to the saccade generator for behavioral execution is 50 ms (Tehovnik et al. 2003). If we assume that the shortest latency from retina to V1 is 30 ms (see: Schiller and Tehovnik 2015) then the shortest fixation time should be about 80 ms (i.e., 50 + 30 ms), which is indicative of an express saccade which is known to be dependent on an intact V1-collicular pathway (Schiller and Tehovnik 2015). As well it is established that when evoking staircase saccades from the colliculus, the fixation time between saccades is as short as ~ 90 ms (see Footnote 1), which we will refer to as falling within the range of express fixations (i.e., 80 to 125 ms). In Figure 2A of Ito, Maldonado et al. (2022) of 19 spontaneously generated saccades made to a complex visual image all fixation durations range from 80 to 300 ms (see Footnote 2 and Fig. 1).

We would suggest that fixation durations of less than 125 ms (see Footnote 3) are too short to do any meaningful visual analysis utilizing the frontal/temporal lobes (see Fig. 15-12 of Schiller and Tehovnik 2015). Clearly, express-saccade durations in overtrained monkeys are not usable for meaningful visual analysis even at the level of V1, which depends on participation from the frontal (and the temporal) lobes. But those who have been studying the frontal/temporal lobes for decades already suspected this (e.g., see Chen and Wise 1995ab; Logothetis et al. 1995; Penfield and Roberts 1966; Squire et al. 2001; Tehovnik, Patel, Tolias et al. 2021).

Finally, we conclude that both mouse and monkey V1 must be studied using freely- moving conditions to permit for volitional control by an animal (Held and Hein 1963; Ito, Maldonado et al. 2022; Xu, Niell et al. 2023): for the mouse the head must be free to orient along with the eyes since the oculomotor range in the mouse is zero, and for the monkey the head can be fixed for visual objects falling within its oculomotor range of 30 degrees, but beyond 30 degrees to head would need to be free (Tehovnik, Hasabegović, Chen 2024; Tehovnik, Patel, Tolias et al. 2021). Remember unlike the mouse with no frontal eye fields (Tehovnik, Hasabegović, Chen 2024), a primate such as a human has these fields for reading the text while the head is immobile (Tehovnik et al. 2021); indeed, reading is compromised after frontal-eye-field damage in humans (Penfield and Roberts 1966).

Footnote 1: See Fig. 15-8A of Schiller and Tehovnik (2015). In the figure each saccade of a staircase sequence was about 20 degrees in amplitude with a duration of 70 ms (Fig. 2C of Chen et al. 2021); since three 20-degree saccades were evoked in 480 ms, this means that the inter-saccadic interval was 90 ms, on average: (480 ms/3 saccades) – 70 ms-saccade-duration.

Footnote 2: The individual durations are in order of occurrence from the study of Ito, Maldonado et al. (2022), Fig. 2A: 120 ms, 200 ms, 150 ms, 80 ms, 120 ms, 120 ms, 150 ms, 300 ms, 200 ms, 100 ms, 200 ms, 150 ms, 200 ms, 150 ms, 150 ms, 200 ms, 120 ms, 100 ms, and 200 ms; also see our Fig 1].

Footnote 3: Express durations occurred for 6 fixations out of 19 in Fig. 2A of Ito, Maldonado et al. 2022, which overlaps with the range of ratios of express to regular saccades in overtrained ‘fixating’ monkeys using a gap of ~ 100 ms (see Fig. 15-12 of Schiller and Tehovnik 2015; also see Fig. 7 of Chen et al. 2021).

Figure 1: The frequency of fixations of a visual image by a monkey as a function of fixation duration. Notice that the fixation durations can be divided into two groups to complement the well-known division between express and regular saccades in primates at a 125 ms dividing line (see Fig. 15-12 of Schiller and Tehovnik 2015). This dividing line can be used to propose a minimal versus a maximal synaptic pathway through the cerebellum (using the thinking of Lisberger 1984) such that: a minimal pathway based on latencies of 80 to 125 ms includes V1, the superior colliculus (Sc), the cerebellar nuclei (Cn), and finally the saccade generator (Sacc Gn); and a maximal pathway based on latencies of 125 to 300 ms includes V1, the frontal and supplementary eye fields FEF/SEF and the superior colliculus Sc, the cerebellar cortex CCx, the cerebellar nuclei Cn, and finally the saccade generator Sacc Gn. The ‘minimal’ pathway has been attributed to mediating ‘unconscious/motor’ vision and the ‘maximal’ pathway subserves largely ‘conscious/declarative’ vision. For complete arguments see Tehovnik, Hasabegović, Chen (2024) & Tehovnik, Patel, Tolias et al. (2021). The data of the graph are from Ito, Maldonado et al. (2022), their Fig. 2A.