Summary of the Special Session &Quot;5 Calories&Quot; Or &Quot;Low&Quot; Calories: How Consumers Use Numerical and Verbal Product Information

Madhubalan Viswanathan, University of Illinois, Urbana-Champaign
[ to cite ]:
Madhubalan Viswanathan (1994) ,"Summary of the Special Session &Quot;5 Calories&Quot; Or &Quot;Low&Quot; Calories: How Consumers Use Numerical and Verbal Product Information", in NA - Advances in Consumer Research Volume 21, eds. Chris T. Allen and Deborah Roedder John, Provo, UT : Association for Consumer Research, Pages: 397-398.

Advances in Consumer Research Volume 21, 1994      Pages 397-398

SUMMARY OF THE SPECIAL SESSION

"5 CALORIES" OR "LOW" CALORIES: HOW CONSUMERS USE NUMERICAL AND VERBAL PRODUCT INFORMATION

Madhubalan Viswanathan, University of Illinois, Urbana-Champaign

The objective of this session was to examine how consumers use numerical and verbal magnitude information, where magnitudes refer to product information such as "120 calories" and "high calories" that convey the location of a product along an attribute. The session brought together papers in the area of numerical and verbal information that relate to several facets of consumer memory and decision making to provide a sense of the range of issues involved in this area of research and to bring out the importance of this hitherto neglected area of research to consumer research and practice. Abstracts of papers are included in the proceedings.

The first paper presented by Terry Childers investigated how numerical and verbal information is represented in memory. Competing perspectives about the nature of memory representations of numerical and verbal magnitudes were tested. An experiment used a learning task followed by a speeded recognition task to test alternate hypotheses. The results suggested that numerical and verbal information may be represented in memory in a verbal-like form.

The second paper presented by Madhubalan Viswanathan examined how consumers encode and use numerical and verbal product information. Across several studies, numerical information was found to be recognized faster and more accurately, to be recalled more exactly, and to require less encoding time, when compared to verbal information for a learning task. However, several advantages for numerical information disappeared for a choice or a judgment task. Further, differences in memory for information as well as the relationship between memory and subsequent judgments were found across tasks.

The third paper by Nancy Artz and Alice Tybout presented by Nancy Artz examined the persuasive implications of numerical and verbal information in advertising. Drawing on differences between numerical and verbal information in terms of precision and evaluativeness, numerical claims were argued to require an additional processing step to interpret the evaluative meaning of a claim, when compared to verbal claims. An experiment demonstrated that a subject's ability to perform the additional processing required by numerical claims has consequences for the persuasive effect of such claims.

J. Russo, the discussant, pointed out that the papers presented dealt with whether numerical versus verbal information is (i) the more natural internal representation, (ii) better remembered after learning, choice, or judgment, and (iii) more persuasive. He presented a broad framework within which these papers could be viewed consisting of three elements; stimuli, processing tasks, and performance tasks with a time delay between processing and performance tasks. Using the framework, he pointed out that the stimuli used in the papers varied in familiarity, in presenting one or both modes of information, and in past knowledge that subjects may have about the stimuli; the processing tasks varied in difficulty as well as anticipated and actual effort; the distracter task used in the time delay varied; and the performance tasks used varied within and across studies. Using the general framework as a basis, he pointed out that there could be multiple factors operating between stimuli and performance tasks and that it is difficult to go from "molar" concepts such as numerical and verbal information to underlying processes. He presented two issues with respect to a future agenda, the first pertaining to context specificity versus generalizability and the second pertaining to simultaneous presentations.

 

REPRESENTATION OF NUMERICAL AND VERBAL INFORMATION IN MEMORY

Terry L. Childers, University of Minnesota

This paper investigated how numerical and verbal information is represented in consumer memory. Competing perspectives about the nature of memory representations of numerical and verbal magnitudes were tested. One perspective tested here was that both numerical and verbal information is represented in memory in an abstract form in terms of the meaning or magnitude conveyed by them. A competing perspective tested was that numerical and verbal magnitudes are represented in their original forms in memory.

The experiment used to test these perspectives involved exposure of subjects to numerical and verbal information describing product attributes during a learning task followed by a recognition task. The mode as well as the meaning or semantic content of the information was manipulated at exposure as well as at recognition. Hypotheses were developed in terms of speed of recognition to test the two perspectives. Specifically, if numerical and verbal information is represented in memory in an abstract form, then recognition speed was predicted to be identical for information at recognition that was similar in meaning to information at exposure. If numerical and verbal information is represented in memory in its original form, then recognition speed was predicted to be identical for information at recognition that was similar in mode to information at exposure. The results suggested that numerical and verbal information may be represented in memory in a verbal-like form. This research provides insights into the nature of representation of numerical and verbal information in memory and has implications for consumer memory and decision making involving these two modes of information.

 

PROCESSING OF NUMERICAL AND VERBAL INFORMATION: IMPLICATIONS FOR CONSUMER MEMORY AND JUDGMENT

Madhubalan Viswanathan, University of Illinois, Urbana-Champaign

This study focused on how consumers encode and use numerical and verbal product information. A framework based on differences in how magnitude information is presented to consumers (i.e., numerical versus verbal modes of presentations) and, differences in how magnitude information is processed by consumers was used to derive hypotheses. Magnitudes presented in numerical versus verbal modes were argued to differ in the degree to which they describe the location of a brand on an attribute in that a verbal magnitude readily conveys its meaning (i.e., the description of the location of a brand along an attribute) whereas a numerical magnitude has to be compared to other information in order to derive meaning from it. Further, the processing of magnitude information was distinguished in terms of either processing a magnitude in terms of its surface features (i.e., surface level processing) or in terms of its meaning (i.e., meaning level processing). This distinction is important for the processing of numerical and verbal magnitudes because differences between numerical and verbal labels in the degree to which the meaning conveyed by them is readily available may lead to different degrees of meaning versus surface level encoding.

Using the framework described above, verbal information was argued to be processed in terms of its meaning to a greater extent than numerical information during learning. During choice or judgment, both types of magnitudes were argued to be processed to a greater extent in terms of their meaning. Hypotheses were derived for differences in encoding, recognition, recall, and judgment between these two modes of information for learning, choice, and judgment tasks. Across several studies, numerical information was found (i) to be recognized faster and more accurately, (ii) to be recalled more exactly, and (iii) to require less encoding time, when compared to verbal information for a learning task. However, several advantages for numerical information disappeared for a choice or a judgment task. Further, a larger degree of recall of numerical information in a verbal form was found following choice or judgment when compared to learning, perhaps because of a greater degree of processing of numerical information in order to make an inference from it in these tasks. Also, a stronger relationship was found following choice or judgment between the evaluative equivalent of numerical information recalled in a verbal (i.e., descriptive) form and subsequent judgments than between the evaluative equivalent of numerical information recalled in a numerical form and subsequent judgments, perhaps because of a greater degree of usage of descriptive information. Implications of this research for the processing of numerical and verbal magnitude information as well as the processing of magnitude information in general were discussed.

 

NUMERICAL AND VERBAL INFORMATION: IMPLICATIONS FOR PERSUASION

Nancy Artz, University of Southern Maine

Alice Tybout, Northwestern University

This research examined the persuasive implications of numerical and verbal information in advertising. Drawing on differences between numerical and verbal information in terms of precision and evaluativeness, numerical claims were argued to require an additional processing step to interpret the evaluative meaning of a claim, when compared to verbal claims. Consequently, when processors do not have the knowledge required to interpret the evaluative meaning of a numerical claim, they may use contextual cues to a greater extent than when they do have the required knowledge. Therefore, judgments will be influenced by contextual cues to different degrees depending on the availability of knowledge to interpret the evaluative meaning of a numerical claim.

An experiment demonstrated that a subject's ability to perform the additional processing required by numerical claims has consequences for the persuasive effect of such claims. An experimental procedure was used where subjects formed impressions of products using numerical or verbal information. Task demands as well as availability of reference information to interpret numerical information were manipulated. Hypotheses relating to the effects of different conditions on consumer judgment were tested. The results suggest that the processing of numerical claims is similar to that of verbal claims when subjects have reference information and sufficient cognitive resources to interpret the evaluative meaning of numeric claims. However, when subjects are unable to interpret the evaluative meaning of numeric claims, differences in the processing of numeric versus verbal claims are found. Specifically, subjects tended to use contextual cues as a basis for forming their evaluative judgments. This research brings out the importance of studying how consumers process numerical and verbal information in terms of their potential effects on persuasion.

----------------------------------------