SDT attempts to describes two main phenomena:

1. Stimulus detection: What is the minimum amount of stimulus energy (loudness, brightness, etc.) that must be present in order for us to detect the presence of the stimulus?
2. Stimulus discrimination: What is the minimum difference in stimulus energy that two stimuli must have for us to be able to tell them apart?

The lowest energy a stimulus can have and still be detected is called the threshold of detection.

There are three main methods for measuring a threshold:

1. Method of limits — Present stimuli in ascending or descending order.


Ascending order — Report when stimulus is first noticed.

Descending order — Report when stimulus first disappears

2. Method of adjustment — Stimulus is continually adjusted upwards or downwards until it is either noticed or disappears.
3. Method of constant adjustment — A set of stimuli are presented in random order multiple times. The percentage of times a subject says they perceived it is the measure.

The smallest difference two stimuli need to have to be discriminable is the just noticeable difference (JND).
 
 
 
 

The JND changes as a function of the size of the stimuli.

What stimulus parameters affect our ability to find a given stimulus (the target) within a group of other stimuli (distractors)?

Two types of search:

1. Feature search — when the target shares no (physical) features with the distractors, such as an X in a field of O’s, then the number of distractors present does not affect how long it takes to determine whether the target is present or not.
2. Conjunction search — When the target differs from the distractors by only one feature, and shares with them at least one more, the amount of time it takes to determine whether the target is present increases as the number of distractors increases.

Triesman claims that when we are conducting a search, we develop a set of feature maps, one for each feature of the objects present, that are used in parallel across the entire search space to identify whether those features are present in the search space or not.

In feature search, since the target shares no features with the distractors, it will immediately activate a corresponding feature map, and thus the target will pop out at the searcher, meaning that the number of distractors will not affect search time.

In conjunction search, since the target shares at least one feature with all of the distractors, we need to find an object where a conjunction of these feature maps is present. We can only search for these conjunctions in one object at a time, which is why the number of distractors affects search time.

Similarity theory — It’s just, in general, harder to discriminate objects that are similar to each other, and thus conjunction search will be harder than feature search.

Guided search — Two stages:

A parallel stage where every item that could possible be a target is highlighted based on whether it shares a feature with the target.

A serial stage where each possible target is evaluated.

In feature search, the parallel stage will only activate the target because it is the only object that has all the features of the target. Thus, the serial stage has no work to do, and the number of distractors has no effect.

In conjunction search, the parallel stage will activate all the distractors that share features with the target, and these would each have to be evaluated in the serial stage, meaning that the number of distractors will affect the search time.

Selective attention is what people colloquially think of when they think of "attention." It is attending to one stimuli while ignoring other competing stimuli.

Three primary theories of selective attention:

1. Filter theory — attention filters out extraneous information
2. Bottleneck theory — attention is a bottleneck that prevents extraneous information from getting through. (Similar to filter theory)
3. Resource theory — attention is a resource-based process, and once the resource is used up, no other information can be attended to.

Basic filter theory (Broadbent)

Filter theories assume that attention acts as to filter out extraneous information, only letting information relevant to the current task get processed.

Filter theory says that at a party, we filter out noise extraneous to our conversation, so that very little of it gets through to be processed and understood.
 
 

Attenuation theory (Triesman)

A modified version of filter theory, it says that rather than filtering out extraneous information, the strength of this information is simply weakened, or attenuated.

Late-filter theory (Deutsch & Deutsch)

This model simply allows for the possibility that all information gets processed at some basic level, but only some of it breaks through to consciousness.

Assumes that "attention" is actually a quantity, or resource, that people have a limited amount of.

Tasks each take a certain amount of attention, and we can attend to multiple tasks only up to the point that we have use up all of our attentional resources.

Example: Driving and conversing with the person next to you vs. Driving and talking on a cell phone.

By definition, since automatic processes are non-conscious in nature, they presumably do not require a large amount of attention to perform. How, then, do they interact with tasks that are attention requiring?

Stroop Effect

The Stroop effect happens because the automatic reading process interferes with the desired task of color naming. Why could this be the case?
 
 

On the other hand, driving with a cell phone is a case where an attention-requiring process interferes with an automatic one. How could this happen? What are the conditions that can cause either the Stroop effect or the increased danger of driving with a cell phone?

Put another way, do automatic processes really not require any attention? If they do require attention, are they truly "non-conscious?"