C. How could you perform the cognitive tasks necessary to achieve your present goals, meet your present needs, without selecting out what input to process?

II. What phenomena about attention does a theory need to account for? - Selective attention and divided attention

A. How does an attention mechanism allow us to select some input for processing and not select other input

a. One possibility is that all of the activity is in the selecting; not selecting is simply the absence of selecting

b. Another possibility is that not selecting actually is an active process too; an act of blocking

(2) which inputs are processed must be determined by characteristics of the input (bottom up)

(1) implies that all input is processed, but differently; some is allowed to be processed further (selected), some not (blocked or ignored)

III. First I will present an account that favors the idea that attending involves actively processing both targets and distractors. Then I will present a different account that favors actively processing the targets, but not the distractors.

IV. What kind of an experimental task is useful for dealing with these questions?

1. A person wears stereo headphones through which separate input can be played (separate channels)

2. To induce attending to one channel, the subject is required to "shadow" a verbal input

a. This allows tests of what the subjects is or is not aware of in the unattended channel

a. People report being unaware of what words, or even what language, was being spoken in the unattended channel

b. They show no memory for words repeated in the unattended channel (possibility they might recognize words even though reporting not being aware of them)

3. These results can be interpreted generally as showing that what gets through the unattended channel is physical characteristics (lower level characteristics), but not meaning (higher level processing).

2. Person will notice their own name (meaning) or other word that has importance to them in the unattended channel

A. About whether selecting is the only active part of the process or whether both selecting and not selecting are both active?

2. Even the un-attended channel is examined at least to separate the physical characteristics from the meaning

B. About whether there is a top down process that at some point processes the input that ends up being not selected?

2. Some process must examine the unattended input to determine whether or not the meaning has any importance to the listener - can only be determined by characteristics of the listener - what is important to one person may not be to another (their name, for instance).

VI. A selective priming account

A. So far I have led you to believe that attentional phenomena can best be accounted for by a mechanism that directs its activity towards both the target and the distractor

B. Now we will consider a mechanism to account for these same phenomena by focusing on only the target, not the distractor - priming the target

1. On the one hand such an account can explain a phenomenon not well explained by the other account - the fact that people can shadow a predictable message better than a more random message

2. But on the other hand - if we attend by "preparing for", or "priming" the relevant detectors, then what's to prevent us from priming multiple detectors? In other words why can't we attend to as many different things as we want to?

VII. A selective priming account’s advantages and disadvantages

A. Advantages - can account for many findings

1. Ability to shadow a particular message - prime detectors for the appropiate ear

2. Cocktail party effect - name is already well primed

3. Predictable message is easier because one part of message primes others - with random message, it's less likely any one part will prime another because they are not related

B. Disadvantage - need a way to account for the limits of ability to divide attention

1. People learn some pretty amazing dual task feats

a. Shadowing one message while typing a visually presented one

b. Nevertheless, there is a limit to how many things we can attend to - and even the amazing feats take practice, not just deciding to do two things at once

VIII. What might prevent us from priming all the detectors we want to prime is some kind of limited resource. Therefore, if we can find evidence for limited resources, then we could explain the phenomena as described in the advantages, but still have an explanation for the limits.

A. Posner and Snyder (1975)

1. Participant's task: Respond "same" or "different" to pairs of letters

a. AA - same

b. AB - different

2. First factor varied - Type of prime

a. Neutral - + then AA or AB

b. Prime - A then AA or AB

c. Misled - G then AA or AB

3. Second factor varied - validity of prime

a. Low validity - could provide data driven prime, but not conceptual prime (if 2 types)

b. High validity - could provide data driven prime and conceptual prime (if 2 types)

B. Results

1. In Low Validity

a. Prime provided a speed advantage over neutral

b. Misled resulted in neither advantage nor cost

2. In High Validity

a. Prime provided even greater speed advantage

b. Misled resulted in just as great a cost

C. Interpretation

1. Supports two types of priming - data driven and conceptual

2. Expectation based priming (conceptual) must draw on a limited resource

a. The data can prime one detector without taking away anything from an unprimed detector (if the physical characteristics of G is primed, A is just not primed, but A is not un primed)

b. Preparing for meaning requires limited resources ( if the meaning of G is primed, then the meaning of A has priming subtracted from it).

D. Implications

1. At least one kind of priming, i.e., MEANING priming, has limited resources

2. Accounts for why we notice the physical characteristics of the unattended channel (male or female voice) - we don't have to draw on a limited resource to prime detectors for physical characteristics

3. Accounts for why we do not attend to meaning in the unnattended channel - we use up all of our meaning priming resources to attend to the meaning of the shadowed message.

A. Has also gotten same results when stimulus being attended was location not shape of stimulus (letters)

2. People can widen or narrow the location attended, but they can't attend to two separate locations (with an unattended area in between)

X. We have been addressing performance, but what about the actual brain activity that produces the performance?

b. Computer mathematically combines them to produce one image with best of both worlds

a. Associated anatomically with working memory! (AWARENESS - AN ASPECT OF CONSCIOUSNESS)

b. Associated with initiation of action (INTENTIONALITY!! ANOTHER ASPECT OF CONSCIOUSNESS)

(4) At worst - neglect, but UNAWARE OF NEGLECT, REPORT EXCUSES INSTEAD OF NEGLECT

A. Evidence for both active process to select distractor and active process to ignore distractor

B. We attend by preparing for (priming) the target, but there are two kinds of priming

C. The performance evidence and the neurological evidence point to the activity of many distinct processes which taken together provide the unitary experience of selective attention