Morton's Logogen Model

Visual

Inputs

Auditory

Contextual

"DO"; "OG"

/d/; /o/; /g/

"I have a pet..."

logogen for "dog"

Examples

Each feature adds to logogen's activation

+

+

100%

Eventually, the logogen for the word reaches its activation threshold, and fires

access to word "dog"

decay (~1 sec)

Word remains accessible for ~1 second while activation decays

Input moves bottom-up (except for regarding contextual input), meaning the logogen model can't account for word superiority.

Logogens act independently and don't compete. Each act as feature counters, gaining activation as inputs that are relevant to its word are received, unaffected by and unaffecting the activation of other logogens.

The logogen model can account for priming effects through its explanation of contextual inputs. Thus, by seeing semantically similar words, you increase the activation of the target logogen, decreasing the time needed to make lexical access to it.

The logogen model is able to explain degradation of stimulus effects, as a lower-quality stimulus is slower to be processed.

Because the activation of words takes about a second to fully decay, if the word is immediately read or heard again, the activation of the associated logogen is higher than the baseline activation, and thus will be quicker to fire than the first time. This explains priming effects.

According to the logogen model, words that are generally more frequently seen or heard adopt a lower threshold for activation than less-frequent words, making them quicker to access. This explains frequency effects.