To Decrease The Strength Of The Stimulus, Which Control Would You Use?

Effigy 1

Direction-bigotry task and random dot motility stimulus. A. Left-correct, direction-of-motion discrimination job. On each trial, the observer fixates a central fixation signal so targets appear to the left and right. After an exponentially distributed random foreperiod, the random dot motion stimulus is presented. Observers view the stimulus until they brand a response, indicating their judgment most the direction of motion by making a saccade to one of the targets. B. Instance of a random-dot move stimulus of variable motility coherence. Stimulus strength is varied past changing the proportion of dots moving coherently in a single management.

Effigy 2

Motion strength affects response times and accurateness. The top panel shows the hateful response time for right responses on a log calibration, and the bottom panel shows the proportion of correct responses. Both graphs are a function of motion strength on a log scale. Error bars represent 1 SE in all figures. Smoothen curves depict the predicted functions from the all-time-fitting proportional-rate improvidence model. Data are from Roitman and Shadlen (2002).

Effigy three

An illustration of a sample path of the accumulation of evidence underlying a perceptual decision. On each trial, evidence in favor of one alternative over another is accumulated as a function of time. For any particular stimulus strength, there is an accumulation of noisy evidence parameterized by the mean rate of accumulation. A decision is made when the process reaches one of the bounds.

An illustration of a sample path of the accumulation of evidence underlying a perceptual decision. On each trial, show in favor of one culling over another is accumulated equally a function of time. For any detail stimulus force, there is an aggregating of noisy evidence parameterized past the mean rate of accumulation. A determination is fabricated when the process reaches 1 of the premises.

Figure 4

An illustration of the relation between the chronometric and psychometric function. On a log-scaled stimulus strength axis, both functions have a fixed shape between upper and lower asymptotes. The sensitivity parameter shifts both functions horizontally in unison.

An analogy of the relation between the chronometric and psychometric function. On a log-scaled stimulus strength axis, both functions have a fixed shape between upper and lower asymptotes. The sensitivity parameter shifts both functions horizontally in unison.

Effigy five

How parameters touch on the chronometric and psychometric functions. A. Chronometric and psychometric functions for three values of the normalized leap A′. Increasing the spring increases the highest value of the chronometric function and decreases the halfway threshold for both functions. B. Chronometric and psychometric functions for three values of sensitivity k. Increasing sensitivity decreases the halfway threshold for both functions. C. Chronometric and psychometric functions for three values of the mean residuum time t. Increasing hateful remainder fourth dimension displaces the chronometric function upward.

Effigy half-dozen

Experiment 1: Response time and accuracy as a function of motion strength. For six observers, each pair of panels shows mean response time for correct responses and proportion of correct responses as a function of motion strength on a log scale. Smooth curves depict the predicted functions from the best-fitting proportional-rate diffusion model.

Experiment 1: Response time and accuracy as a function of motion strength. For six observers, each pair of panels shows mean response time for correct responses and proportion of correct responses as a function of movement strength on a log scale. Smooth curves depict the predicted functions from the best-fitting proportional-rate diffusion model.

Figure 7

Experiment 2: Response time and accuracy as a function of motion strength and speed instruction. Observers are shown in separate columns. Speed instructions had large effects on response time for low motion strengths and little effect for high motion strengths.

Experiment 2: Response fourth dimension and accuracy equally a role of move force and speed instruction. Observers are shown in carve up columns. Speed instructions had big furnishings on response time for depression motion strengths and lilliputian effect for loftier motion strengths.

Figure eight

Experiment 2: Effect of speed instruction on parameters of proportional-rate diffusion model. A. Speed instruction effects on bound. The bound increases with the increasing time of the speed instruction. B. Speed instruction effects on sensitivity. C. Speed instruction effects on mean residual time. The primary effect of the speed instruction is on the bound.

Experiment 2: Result of speed pedagogy on parameters of proportional-rate diffusion model. A. Speed instruction effects on jump. The bound increases with the increasing time of the speed instruction. B. Speed instruction effects on sensitivity. C. Speed instruction effects on mean residual time. The primary effect of the speed didactics is on the bound.

Figure nine

p ]Experiment 3: The proportional-rate diffusion model can account for response times that obey Piéron's Law. Smooth curves depict the predictions of the best-fitting proportional-rate diffusion model and dashed curves depict the predictions of Piéron's Law.

p ]Experiment 3: The proportional-rate diffusion model can business relationship for response times that obey Piéron'due south Law. Smoothen curves depict the predictions of the best-fitting proportional-rate diffusion model and dashed curves describe the predictions of Piéron's Police.

Figure 10

Experiment 4: Response time and accuracy as a function of motion strength for eye and finger movements. Observers are shown in separate columns. The proportional-rate diffusion model fits both kinds of responses.

Figure xi

Experiment 5: Discrimination: Response time and accuracy as a function of the contrast increment. The power-rate diffusion model fits this contrast discrimination task with an exponent of about 1.2.

Experiment 5: Discrimination: Response time and accuracy equally a function of the contrast increment. The ability-charge per unit diffusion model fits this contrast discrimination task with an exponent of almost 1.2.

Figure 12

Experiment 5: Detection: Response time and accuracy as a function of contrast in a detection task. For detection, the power-rate diffusion model fits the data with an exponent of about 2.0.

Experiment 5: Detection: Response fourth dimension and accuracy every bit a function of contrast in a detection job. For detection, the power-rate improvidence model fits the data with an exponent of about ii.0.

Figure xiii

Summary of threshold ratio estimates from all experiments.

Figure 14

Relationship between accuracy threshold and estimated decision time in Experiment 2. Accuracy threshold decreases with the square root of the estimated decision time at accuracy threshold. The diagonal line has a log-log slope of −1, which follows Bloch's law.

Relationship between accuracy threshold and estimated decision fourth dimension in Experiment 2. Accuracy threshold decreases with the square root of the estimated determination fourth dimension at accuracy threshold. The diagonal line has a log-log slope of −1, which follows Bloch's law.

Figure 15

Comparison of error and correct response time. Mean error response fourth dimension is plotted confronting mean correct response time. A. Results for six observers from Experiment one. Each point corresponds to the error and right mean response times for a given motion force in a detail observer. Information from the two everyman nonzero motion strengths are shown. The diagonal line indicates equal response times. B. Like plot for observer JP from the other experiments.

Table 1

Gaze precision and percentage rejected trials for all experiments. SD = standard deviation, Disc. = discrimination, and Det. =detection.

Table ii

Experiment ane: Parameter values for proportional-rate diffusion model. L = likelihood; t _R in seconds.

Table 3

Experiment 2: Parameter values for uncoupled model. Speed instructions and tR in seconds. The χⁱⁱ is based on a likeli-hood ratio test comparing this uncoupled model to the propor-tional-rate diffusion model. * p <.05, ** p< .01.

Table four

Experiment ii: Parameter values for ability-rate improvidence model. Speed didactics and tR in seconds. The χ² is based on a likelihood ratio test comparing this power-rate diffusion model to the proportional-rate improvidence model. *p < .05, ** p < .01, *** p <.001.

Table 5

Experiment four: Parameter values for proportional-rate improvidence model. L = likelihood; t _R in seconds.

Table 6

Experiment 5: Parameter values for power-rate diffusion model. Fifty = likelihood; t _R in seconds.

Table A1

Summary of notations used in the 1.