Two-pulse discrimination and rapid light adaptation: complex effects on temporal resolution and a new visual temporal illusion

A Role for Bottom-Up Alpha Oscillations in Temporal Integration

Neural oscillations in the 8-12 Hz alpha band are thought to represent top-down inhibitory control and to influence temporal resolution: Individuals with faster peak frequencies segregate stimuli appearing closer in time. Recently, this theory has been challenged. Here, we investigate a special case in which alpha does not correlate with temporal resolution: when stimuli are presented amidst strong visual drive. Based on findings regarding alpha rhythmogenesis and wave spatial propagation, we suggest that stimulus-induced, bottom-up alpha oscillations play a role in temporal integration. We propose a theoretical model, informed by visual persistence, lateral inhibition, and network refractory periods, and simulate physiologically plausible scenarios of the interaction between bottom-up alpha and the temporal segregation. Our simulations reveal that different features of oscillations, including frequency, phase, and power, can influence temporal perception and provide a theoretically informed starting point for future empirical studies.

Theoretical and Technical Issues Concerning the Measurement of Alpha Frequency and the Application of Signal Detection Theory: Comment on Buergers and Noppeney (2022)

Classical and recent evidence has suggested that alpha oscillations play a critical role in temporally discriminating or binding successively presented items. Challenging this view, Buergers and Noppeney [Buergers, S., & Noppeney, U. The role of alpha oscillations in temporal binding within and across the senses. Nature Human Behaviour, 6, 732-742, 2022] found that by combining EEG, psychophysics, and signal detection theory, neither prestimulus nor resting-state alpha frequency influences perceptual sensitivity and bias in the temporal binding task. We propose the following four points that should be considered when interpreting the role of alpha oscillations, and especially their frequency, on perceptual temporal binding: (1) Multiple alpha components can be contaminated in conventional EEG analysis; (2) the effect of alpha frequency on perception will interact with alpha power; (3) prestimulus and resting-state alpha frequency can be different from poststimulus alpha frequency, which is the frequency during temporal binding and should be more directly related to temporal binding; and (4) when applying signal detection theory under the assumption of equal variance, the assumption is often incomplete and can be problematic (e.g., the magnitude relationships between individuals in parametric sensitivity may change when converted into nonparametric sensitivity). Future directions, including solutions to each of the issues, are discussed.

Sensory Drive Modifies Brain Dynamics and the Temporal Integration Window

Perception is suggested to occur in discrete temporal windows, clocked by cycles of neural oscillations. An important testable prediction of this theory is that individuals' peak frequencies of oscillations should correlate with their ability to segregate the appearance of two successive stimuli. An influential study tested this prediction and showed that individual peak frequency of spontaneously occurring alpha (8-12 Hz) correlated with the temporal segregation threshold between two successive flashes of light [Samaha, J., & Postle, B. R. The speed of alpha-band oscillations predicts the temporal resolution of visual perception. Current Biology, 25, 2985-2990, 2015]. However, these findings were recently challenged [Buergers, S., & Noppeney, U. The role of alpha oscillations in temporal binding within and across the senses. Nature Human Behaviour, 6, 732-742, 2022]. To advance our understanding of the link between oscillations and temporal segregation, we devised a novel experimental approach. Rather than relying entirely on spontaneous brain dynamics, we presented a visual grating before the flash stimuli that is known to induce continuous oscillations in the gamma band (45-65 Hz). By manipulating the contrast of the grating, we found that high contrast induces a stronger gamma response and a shorter temporal segregation threshold, compared to low-contrast trials. In addition, we used a novel tool to characterize sustained oscillations and found that, for half of the participants, both the low- and high-contrast gratings were accompanied by a sustained and phase-locked alpha oscillation. These participants tended to have longer temporal segregation thresholds. Our results suggest that visual stimulus drive, reflected by oscillations in specific bands, is related to the temporal resolution of visual perception.

The Peripheral Flicker Illusion

A new illusion is reported. A visual object suddenly appearing on a red background sometimes causes an impression of flicker or double flash. In Experiment 1, a red, green, or blue object was presented on a red, green, blue, or gray background. Participants evaluated the illusion strength in reference to the physical flicker of a gray object presented in central vision. The results show that the green or blue object presented on the red background caused the illusion. In Experiment 2, the effect of retinal eccentricity on the illusion was tested. The results showed that the illusion was weak in central vision but became stronger as the retinal eccentricity of the objects' presentation increased. In Experiment 3, optimal luminance conditions for the illusion were explored with the green and blue objects. The illusion was strong when object luminance was lower than background luminance and the optimal luminance for the blue object was lower than that for the green object. We propose a tentative theory for the illusion and discuss its cause.

Testing a computational model of light-adaptation dynamics

Experiments from the periodic and aperiodic traditions were used to guide the development of a quantitatively valid model of light adaptation dynamics. Temporal contrast sensitivity data were collected over a range of 3 log units of mean luminance for sinusoids of 2 to 50 Hz. Probe thresholds on flashed backgrounds were collected over a range of stimulus-onset asynchronies and background intensities from 0.1 to 1000 td. All experiments were performed foveally in the photopic range and used a consistent stimulus paradigm and psychophysical method. The resulting model represents a merging of elements from both traditions, and consists of a frequency-dependent front-end followed by a subtractive process and static nonlinearity.

Anatomy of a flash. 1. Two-peak masking and a temporal filling-in

A modified paradigm of Crawford masking was used to link masking to brightness fluctuation, as distinct from flash brightness. Thresholds were measured for a 10 ms incremental pulse (the 'probe') presented before, during, or after a 500 ms pulse (the 'flash'). Both pulses were spatially coextensive with the background field, thus the criterion for probe detection was purely temporal. The flash occurred either in the tested eye, the opposite eye, or in both eyes. In all conditions, masking was strongly bimodal: thresholds peaked near flash onset and flash offset. The flash was perceived as a unitary event. Bimodal masking is attributed to cortical on-and off-effects, as (i) dichoptic masking was strong and (ii) the same incremental probe was masked by either incremental or decremental flashes. Strikingly, monocular probe thresholds were about equally elevated by binocular as by monocular flashes, although the binocular flashes were brighter. Therefore, some monocular features can be preserved in the larger net binocular response. A general conclusion is that masking depends on the same transient neural responses that bring about a brightness fluctuation, whereas the appearance of the flash as a single event, a unitary change of brightness, depends on a different mechanism, perhaps a sustained response that performs a temporal filling-in.

Two pulses seen as three flashes: a superposition analysis

If a single brief light pulse follows the offset of a light field by 0.1-0.3 sec, the pulse is seen as a double flash. This "double flash effect" is a suprathreshold phenomenon: the pulse must exceed detection threshold by 10 times or more for this temporal illusion to occur. A special case of this effect is demonstrated here: two brief, high-luminance pulses separated by 0.1 sec appear as three flashes. In a superposition analysis, hypothetical impulse response functions were added together with various delays to model flash perception. A biphasic impulse response (congruent with threshold flicker and pulse sensitivity) fails to predict perception of three flashes from two pulses. The analysis instead suggests that the visual response to a suprathreshold pulse has several alternating phases of excitation and inhibition.

On and off systems in human vision

Three experiments examined the interaction of On and Off responses that were produced by sudden increments and decrements in luminance. All three experiments utilized a masking technique that required observers to detect a signal in a masking field. The mask was produced by brightening or dimming a field of dots, and the signal consisted of the addition or subtraction of a dot. Experiment 1 showed that detection of the signal-dot was more difficult when the luminance of the signal and the mask changed in the same direction (e.g. a new dot added to the field of dots that were being brightened) than when luminance changed in different directions. When the amplitude of signal and mask was varied parametrically (Experiments 2 and 3), accuracy increased with the ratio of amplitudes of signal and mask. But at any given ratio, the signal was more difficult to detect when signal and mask were of the same sign. The greater difficulty encountered in detecting a signal in the context of a "like" mask is ascribed to greater interference between signal and mask when they share the channel.

Improvements in visual performance following a pulsed field of light: a test of the equivalent-background principle

The offset of a pulsed conditioning field of light has recently been shown to produce enhancements of temporal resolution and brightness discrimination. These enhancements are similar to those that are due to imposing a high level of light adaptation on the visual system. Here we analyze whether there is a true equivalence of adaptive state between some level of steady light adaptation and the offset of a conditioning field. The enhancements of visual function at field offset were replicated by using a suprathreshold two-pulse discrimination task and a task requiring detection of an incremental probe stimulus superimposed upon a suprathreshold pulse. These effects are shown to be qualitatively but not quantitatively similar to those produced by an equivalent background selected on the basis of its ability to raise threshold to the same degree as field offset. We conclude that the equivalent-background principle cannot be supported for our measures of foveal visual function.