Psychment Departology

It is hard within the interactive activation model to account for these strings being misread as "PSYCHOLOGY DEPARTMENT". Experimental evidence that the relative positions of words are important in reading was reported by McClelland and Mozer (1986). They found that pairs of words such as LINK and MINE were sometimes misread as LINE and MINK.

There are some limitations with the model put forward by McClelland and Rumelhart (1981). The most obvious one is that it was only designed to account for performance on four-letter words written in capital letters, although it could probably be applied to longer words.

High-frequency or common words are more readily recognised than low-frequency or rare words. This can be explained by assuming either that stronger connections are formed between the letter and word units of high-frequency words, or that high-frequency words have a higher resting level of activation. It follows that there should be a larger word superiority effect for high-frequency words than for low-frequency words due to more top-down activation from the word level to the letter level. However, the size of the word superiority effect is the same with high- and low-frequency words (Gunther, Gfoerer, & Weiss, 1984).

The model proposed by McClelland and Rumelhart (1981) assumes that lexical access is determined by visual information. However, there has been much controversy on this issue. Frost (1998) claimed that phonological coding is nearly always used prior to lexical access (see later in the chapter).

Developments of the model

The original interactive activation model predicted accuracy of word recognition, but could not predict the speed of word reading. This limitation was addressed by Grainger and Segui (1990) and Jacobs and Grainger (1992). They modified the model so that responses were made when activation at the word level reached a variable threshold of activation. With this addition to the model, Jacobs and Grainger (1992) simulated the lexical decision times of human participants. Grainger and Segui (1990) assumed that high-frequency words have a lower activation threshold than low-frequency words. They focused particularly on lexical decision times to low-frequency words (e.g., BLUR) having a similar spelling to a high-frequency word (e.g., BLUE). They predicted (and found) that lexical decision times were slowed down, presumably because activation of the incorrect high-frequency word inhibited activation of the correct low-frequency word.

McClelland (1993) pointed out that the original interactive activation model was a deterministic one, meaning that any given input would always produce the same output. This contrasts with human performance, which is somewhat variable. Accordingly, McClelland (1993) developed the model by including variable or stochastic processes within it. This permitted the model to simulate the response distributions of human participants given various word-recognition tasks.

Stop Anxiety Attacks

Stop Anxiety Attacks

Here's How You Could End Anxiety and Panic Attacks For Good Prevent Anxiety in Your Golden Years Without Harmful Prescription Drugs. If You Give Me 15 minutes, I Will Show You a Breakthrough That Will Change The Way You Think About Anxiety and Panic Attacks Forever! If you are still suffering because your doctor can't help you, here's some great news...!

Get My Free Ebook


Post a comment