Search

Divided attention has little effect for simple tasks, such as luminance detection, but it has large effects for complex tasks, such as semantic categorization of masked words. Here, we asked whether the semantic categorization of visual objects shows divided attention effects as large as those observed for words, or as small as those observed for simple feature judgments. Using a dual-task paradigm with nameable object stimuli, performance was compared with the predictions of serial and parallel models. At the extreme, parallel processes with unlimited capacity predict no effect of divided attention; alternatively, an all-or-none serial process makes two predictions: a large divided attention effect (lower accuracy for dual-task trials, compared to single-task trials) and a negative response correlation in dual-task trials (a given response is more likely to be incorrect when the response about the other stimulus is correct). These predictions were tested in two experiments examining object judgments. In both experiments, there was a large divided attention effect and a small negative correlation in responses. The magnitude of these effects was larger than for simple features, but smaller than for words. These effects were consistent with serial models, and rule out some but not all parallel models. More broadly, the results help establish one of the first examples of likely serial processing in perception.

It has been known for decades that haplorrhine primates and other non-mammalian foveate vertebrates explore visual scenes by making fast eye movements that allow them to shift the image of an object of interest into the fovea. High spatial resolution accommodated by the fovea can be used to scrutinize the object of interest during a period of steady foveal fixation of the object’s image, interrupted only briefly by different types of miniature eye movements, too small to jeopardize the image position within the confines of the fovea. In this dissertation I address a peculiar aspect of the object fixation of macaque monkeys observed under conditions of full darkness. Macaque monkeys exhibit an upward shift of gaze (for short, ‘upshift’) when asked to fixate a tiny fixation spot in an otherwise dark environment. Given the advantages of foveal vision, the upshift that moves the fovea away from the object of interest, the fixation point, must appear paradoxical. The upshift was first described by Snodderly (1987) in rhesus monkeys and rediscovered and investigated in detail by Barash et al. (1998) who studied it in cynomolgus macaques. Barash et al. (1998) demonstrated that the upshift depended solely on the background luminance and not on the contrast between target and background. Barash and coworkers could also show that the upshift increased with training on the fixation and typically started within seconds after turning the background from bright to dark. The view that the upshift was a hitherto undescribed illumination-dependent fixation offset not related to known features of the systems controlling saccades and fixation, was put into a question by Goffart et al. (2006). This group claimed that the upshift is nothing but a luminance dependent disbalance of the saccadic system for the vertical. However, based on this hypothesis, the upshift should be abolished soon after the onset of fixation. To critically test this hypothesis, we carried out a first study in which we trained rhesus monkeys to fixate small targets on the screen. The monkeys fixated a target in two conditions: with bright or dark background. We investigated the time course of the difference between fixation in darkness and fixation in bright conditions that is the actual upshift. We showed that dark-background-dependent upshift persisted during at least during two seconds of fixation. Additionally, the size of the upshift depended on fixation location: fixation in the lower visual field resulted in larger upshifts whereas fixation in the upper visual field demonstrated smaller upshifts. These results clearly indicated that the upshift cannot merely be a consequence of a dysbalanced saccadic system causing hypermetric upward and hypometric downward saccades as both should be corrected within a few milliseconds, ultimately establishing foveal fixation. Anecdotal observations seemed to suggest that the upshift might differ between monkeys. In an attempt to consolidate this impression and, moreover, to identify a cause explaining eventual differences, we embarked on study of a very large sample of 14 monkeys from two species. We tested the monkeys in the same task that we had used in the preceding study. The results were very clear: all monkeys in our sample had upshift and the monkeys lacked systematic horizontal deviation during fixation in darkness. However, the monkeys tested exhibited substantial differences as to the size of the upshift. The monkeys in our large sample differed by the level of ‘habituation’: dark habituated (monkeys that had been previously trained in tasks in full darkness with small bright stimuli) and bright-habituated (monkeys that had been trained in tasks with large bright stimuli without control for full darkness of experimental setup). We showed that the size of the upshift largely reflects the extent to which a monkey is habituated to work in the dark. Dark habituated monkeys with mostly belonged to 2 the group with higher upshift whereas bright habituated monkeys were very likely to demonstrate lower upshift. Species differences (cynomolgus vs. rhesus macaques) were not found. In seeking to explain the upshift, we resorted to the geometry of the rod and cone densities, which constitute a hard bound for the resolution of the percept. Cones peak in the fovea; extrafoveally, cone density decreases as eccentricity increases. Rods are absent in the foveola. On going dorsally from the foveola, rod density increases, until reaching a peak in a location called dorsal rod peak, or rod hotspot (Packer et al., 1989; Wickler and Rakic, 1990; Wickler et al., 1990). We therefore started with the hypothesis that if any retinal location replaces the fovea in scotopic vision, it would be the rod hotspot. We therefore expected that the vertical component of the line of gaze would be distributed bimodally, one mode reflecting the fovea and the other the rod hotspot. However, the results did not corroborate this hypothesis. Upshift height varied a lot, from monkey to monkey, from condition to condition, even within session. Eventually we were led to an alternative hypothesis, that of a scotopic band. We now suggest that at any epoch during scotopic vision there might be a scotopic center, located dorsally to the fovea. The scotopic center replaces the fovea. During fixations, target images are projected on the scotopic center, not on the fovea. Saccadic trajectories show that saccades shift the target’s image directly to the scotopic center. Therefore, not all saccades foveate. Scotopic saccades do not foveate. The relative weight of the scotopic center in the evolution of saccades remains open. In any case, scotopic saccades come with their own sensorimotor transformations, as do scotopic fixations. Scotopic center is only the beginning of the scotopic analogy for high-acuity vision. Unlike the close, tight fovea of photopic vision, the scotopic center moves along a line extending dorsally from immediately dorsal to the fovea. We call this the scotopic band. The presence of the band makes scotopic sensorimotor transformations more complex than photopic because they rely on a parameter – the scotopic band setting, the current location of the scotopic center on the band. Scotopic band is set primarily according to the ambient light, reflected in the laboratory as background luminance. Increasing luminance results in lower upshift, or, equivalently, more ventral setting. However, other factors heavily influence the scotopic setting too. To mention but 2: habituation to darkness increases the upshift, that is, makes the setting more dorsal; and so do also threshold task conditions. Thus, we suggest that ultimately, the scotopic band setting reflects the statistics of the scene and the monkey’s task. The computational needs change gradually, and so do the anatomy (and physiology). Therefore, scotopic band setting is not limited to the two endpoints but occupy the points in between.

Besides the fundamental competition between the top-down and bottom-up approaches in nanotechnology, there are some basic aspects for organizing structures and functions at the molecular level. The recent challenges to the development of nanotechnology are marked by a group of general requirements: selection of suited building units, overcoming the restrictions of planar technology, shrinking of nanofabrication facilities, sustainable production and management of life cycles, organization of autonomy and communication at the nano-level, and the optimization of power consumption and energy management. Looking at the natural principles in the construction, synthesis, and function of proteins helps in understanding the principal differences between the currently applied technologies and the characteristics of biomolecular mechanisms in cells. This view allows formulating seven basic rules to meet the general requirements for future developments in molecular nanotechnology.

We introduce a novel method capable of dissecting the succession of processing stages underlying mental arithmetic, thus revealing how two numbers are transformed into a third. We asked adults to point to the result of single-digit additions and subtractions on a number line, while their finger trajectory was constantly monitored. We found that the two operands are processed serially: the finger first points toward the larger operand, then slowly veers toward the correct result. This slow deviation unfolds proportionally to the size of the smaller operand, in both additions and subtractions. We also observed a transient operator effect: a plus sign attracted the finger to the right and a minus sign to the left and a transient activation of the absolute value of the subtrahend. These findings support a model whereby addition and subtraction are computed by a stepwise displacement on the mental number line, starting with the larger number and incrementally adding or subtracting the smaller number.

Considerable evidence has accrued on the role of paradigms as both theoretical and cognitive structures regimenting the way words are processed and acquired. The evidence supports a view of the lexicon as an emergent integrative system, where word forms are concurrently and competitively stored as repeatedly successful processing patterns, and on-line processing crucially depends on the internal organisation of stored patterns. In spite of converging evidence in this direction, little efforts have been put so far into providing detailed, algorithmic models of the interaction between lexical token frequency, paradigm frequency, and paradigm regularity in word processing and acquisition. Here we propose a neuro-computational account of the frequency/regularity interaction, and discuss some of its theoretical implications by analysing experimental results in the computational framework of Temporal Self-Organising Maps. Detailed quantitative analysis shows that the model provides a unitary explanatory framework bringing together insights from neighbour family effects on word recognition and production, evidence from family size effects in serial lexical access and paradigm-based dynamics in lexical acquisition.

In this work we study a Hebbian neural network, where neurons are arranged according to a hierarchical architecture such that their couplings scale with their reciprocal distance. As a full statistical mechanics solution is not yet available, after a streamlined introduction to the state of the art via that route, the problem is consistently approached through signal- to-noise technique and extensive numerical simulations. Focusing on the low-storage regime, where the amount of stored patterns grows at most logarithmical with the system size, we prove that these non-mean-field Hopfield-like networks display a richer phase diagram than their classical counterparts. In particular, these networks are able to perform serial processing (i.e. retrieve one pattern at a time through a complete rearrangement of the whole ensemble of neurons) as well as parallel processing (i.e. retrieve several patterns simultaneously, delegating the management of diff erent patterns to diverse communities that build network). The tune between the two regimes is given by the rate of the coupling decay and by the level of noise affecting the system. The price to pay for those remarkable capabilities lies in a network's capacity smaller than the mean field counterpart, thus yielding a new budget principle: the wider the multitasking capabilities, the lower the network load and viceversa. This may have important implications in our understanding of biological complexity.

The interaction between length and lexical status is one of the key findings used in support of models of reading aloud that postulate a serial process in the orthography-to-phonology translation (B. S. Weekes, 1997). However, proponents of parallel models argue that this effect arises in peripheral visual or articulatory processes. The authors addressed this possibility using the special characteristics of the Serbian and Japanese writing systems. Experiment 1 examined length effects in Serbian when participants were biased to interpret phonologically bivalent stimuli in the alphabet in which they are words or in the alphabet in which they are nonwords (i.e., the visual characteristics of stimuli were held constant across lexical status). Experiment 2 examined length effects in Japanese kana when words were presented in the kana script in which they usually appear or in the script in which they do not normally appear (i.e., the phonological characteristics of stimuli were held constant across lexical status). Results in both cases showed a larger length effect when stimuli were treated as nonwords and thus offered strong support to models of reading aloud that postulate a serial component.

We designed two extensions of Saslow's well-known gap and overlap conditions that require increased voluntary effort because of the progressive elimination of target onsets and fixation point offsets, and obtained repeatable data obeying simple numerical relations. For each of the five stimulus lighting conditions, saccadic latency was measured as a function of the retinal eccentricity or displacement of the target. Latencies were fitted by a serial processing model in which the visually guided minimum tracking latency VGL min is supplemented by two types of delay, dubbed `unlock' and `target', that can be either short or long (`direct' or `indirect'), depending on the conditions. There are two findings: (1) The model has utility. The rank order of saccadic latencies for the five stimulus lighting conditions was constant across all subjects, sessions and eccentricities in the range 7.5 ′ –6° left or right. For pooled data, and the saccadic latency plateau (1–6°), the model was also within ±3 ms of the mean latencies. (2) Latencies of tiny saccades to intrafoveolar stimulation (7.5–45 ′ ) were invariably long in all five stimulus conditions. One factor here must be the experimentally measured local prolongation of VGL min .

Does recognizing the transformations that gave rise to an object’s retinal image contribute to early object recognition? It might, because finding a partially occluded object among similar objects that are not occluded is more difficult than finding an object that has the same retinal image shape without evident occlusion. If this is because the occlusion is recognized as such, we might see something similar for other transformations. We confirmed that it is difficult to find a cookie with a section missing when this was the result of occlusion. It is not more difficult to find a cookie from which a piece has been bitten off than to find one that was baked in a similar shape. On the contrary, the bite marks help detect the bitten cookie. Thus, biting off a part of a cookie has very different effects on visual search than occluding part of it. These findings do not support the idea that observers rapidly and automatically compensate for the ways in which objects’ shapes are transformed to give rise to the objects’ retinal images. They are easy to explain in terms of detecting characteristic features in the retinal image that such transformations may hide or create.