The dynamics of information processing: length of the message and interest versus response time
What is the relation, if any, between response time and both the number of letters in the text element, and the interest in the text element, respectively? (Picture elements do not belong here for the obvious reason that they have no text length.) To the degree that these two segments differ in the way they approach the type of credit card information offered, do they also differ in the way they process the information? The impact value for each element and the response time is known, as is the length of the message. A simple equation can be created relating response time to text length and to interest as follows:
Response time =Constant + k_{1}(Number of letters in text) + k_{2}(Interest in text)
Are the models the same for each of the two segments, and how do they compare to the model developed for the total panel?
Table 6 How the number of letters and the interest in the element ‘drive’ response time for total panel and for the two segments


(stuff) 
(lifestyle) 
Additive constant 
7.54 
8.16 
8.91 
Letters (ki) 
0.06 
0.05 
0.07 
Interest (k_{2}) 
0.05 
0.24 
0.19 
R^{2} 
0.38 
0.21 
0.30 
Table 6 shows the results. The coefficient k_{1} for text length is the same for the total panel and the two segments. This means that the longer the text, the longer the processing time. What is unexpected, however, is the difference in sign and magnitude for the impact coefficient. For Segment 1 customers (stuff) the coefficient k_{1} is —0.24, meaning that increases in interest (positive impact) decrease the response time. That is, if the Segment 1 customer is interested, the response is quicker. The Segment 1 customer responds quickly to what he or she likes. In contrast, the Segment 2 customer (lifestyle) responds more slowly to elements that he or she likes. The coefficient k_{2} is 0.19, meaning that if the customer likes the element then the customer will read it more slowly. (Keep in mind that the coefficient refers to tenths of seconds, so a positive coefficient means a slower response time, corresponding to more tenths of seconds.) From a management perspective this means that interesting messages are processed differently by the two segments, and that the two segments have different strategies for responding.