HomeIntroducing CybercultureBook ReviewsCourses in CybercultureEvents and ConferencesFeatured LinksAbout RCCS

View All Books

The Media Equation. How People Treat Computers, Television, and New Media Like Real People and Places

Author: Byron Reeves, Clifford Nass
Publisher: Stanford, CA: CSLI Publications, 1996
Review Published: May 2001

 REVIEW 1: Mikael Jakobsson

This book was first published in 1996 and deals with issues of media and communication. So why is it reviewed as a book of the month at the Resource Center for Cyberculture Studies? The answer is simple: I believe this book to be of fundamental importance to most people engaged in cyberculture studies, especially those who do not themselves directly address issues pertaining to the medium. The book offers an important insight into the human mind that is just as topical today as it was five years ago, and that we cannot afford to oversee. Starting to get interested? Good, then let us look at what the book is actually about.

The media equation looks like this: media = real life. What this means is that the authors claim that people treat computers, television, and new media in general like real (and physically present) people and real places. They also claim that this phenomenon is not rare, inconsequential, or possible to correct with age, education or thought. It happens to us all, it happens often, and it will not go away. To prove their claim the authors put forward accounts of a substantial amount of experiments performed by them and their associates over a number of years. All experiments follow the same pattern. They take a traditional experiment on social interaction concerning manners, personality, emotion, social roles etc. and set it up, but they substitute a person or the environment with media.

The seven steps of testing the media equation:

    1. Pick a social science finding about how people respond to each other or to the natural environment.
    2. Find the place in the report where a social or natural rule is summarized.
    3. Cross out the word "person" or "environment" in the studies, and substitute media.
    4. Find the part of the report that describes how the rule was tested.
    5. Cross out, again one of the people or the description of the environment, and substitute media.
    6. Run the experiment.
    7. Draw implications.

Let us take a closer look at one of these experiments. In this case, the assertion tested was: "When a computer asks a user about itself, the user will give more positive responses than when a different computer asks the same questions." (In this case the original phrasing would have been something similar to: "When a person asks another person about himself or herself, the other person will give more positive responses than when a different person asks the same questions.") Twenty-two people with extensive computer experience were asked to use a computer to learn about various topics. The only thing the computer did was to display text. At the end of the session the computer praised its own performance. The participants were then asked to evaluate the computer they were using in terms of how they felt about it and how well they thought it had performed. Half of the participants used the same computer for this evaluation. The other half used a different computer located at the other end of the room. The participants who used the same computer gave significantly more positive responses than the others did. In fact, compared to similar experiments with humans, the computers got the same treatment.

This is the first experiment reported in the book; it is part of the Media and Manners section. The following sections are Media and Personality; Media and Emotion; Media and Social Roles; and Media and Form. Since this book is an overview of a large body of scientific work, and the intended audience of the book ranges beyond the inner circle of research buffs, the authors never go into figures regarding the statistical strength of their results and other details. For that kind of information you will have to go to the underlying articles published in scientific journals, e.g. Nass, Steuer & Tauber (1994).

This is how they summarize the results of their experiments: "Media are treated politely, they can invade our body space, they can have personalities that match our own, they can be a teammate, and they can elicit gender stereotypes. Media can evoke emotional responses, demand attention, threaten us, influence memories, and change ideas of what is natural. Media are full participants in our social and natural world" (251). The last sentence is an especially crucial finding. It abolishes the idea of the optional and conscious willing suspension of disbelief for a natural and automatic response to media. We have to make a conscious effort not to believe that the computer talking to us is not someone to be thought of as a human.

The authors claim that the media richness is an insignificant factor for the media equation, i.e. the results of the above mentioned experiment are the same independent of how human-like a computer interface is. They do not, however, examine this matter very closely and more research has been made in this area since the book was published, e.g. Bengtsson (1999). Without going into details, their claim still seems to hold. Another question that came up when I read the book was whether the users actually felt that they were interacting with the computer or rather with the programmer making the program. This has also been further investigated in Shaym & Nass (2000). And again, their claim has not been proven wrong.

While their case for the media equation is very convincing, I am not as sure about their explanation of this phenomenon. They try to explain our weak ability to discern between media and reality by arguing that our brain functions were shaped in an unmediated environment and that our ancient brains cannot cope properly with new media, that our brains have not had the chance to adjust to today's technology and therefore instinctively misinterprets it. I find their argumentation weak at a number of points. One is where they relate to an occasion when a ventriloquist's puppet was questioned by the U.S. Congress (!), and the senators thought about the hand puppet as an autonomous individual. (They asked the ventriloquist if she agreed with the puppet.)

As I see it, this has nothing to do with modern media. Puppets have been around more or less forever. And is it not true that we tend to treat cats, dogs, and sometimes even flowers as human actors as well, despite the fact that they have been part of our natural environment for a very long time? Instead of being a phenomenon directly related to new media, I think it is a residue from the preoperational stage (years 2-6) of our cognitive development where we tend to believe that things around us are alive (Piaget, 1929). Although we gradually grow out of this conception, I believe that we have a natural tendency for animism that never quite goes away. Modern media may emphasize this kind of irrational behavior, but on the whole I think their explanation is too narrow.

Out of their results they also try to draw conclusions regarding how to treat and design media in general and computer applications in particular, but the design tips that they give are not as exciting as the results of their experiments and the occasional design implications are not compiled into any kind of framework for media design.

As for the literary qualities of the book, I think the less said the better. It was very exciting to read thanks to its contents but since every chapter follows the exact same formula with an introductory remark based on a researcher's everyday life experiences, stating of the rules, and a brief description of the experiments followed by some conclusions. It got a bit repetitive towards the end. My final verdict is, however, that this book contains an important insight into human interaction with media and that the authors are successful in conveying this message to the reader. The book can and should be used both in research and education within a number of scientific areas but is perhaps especially relevant for people engaged in cyberculture studies, which frequently deals with real humans interacting through new media.

Bengtsson, Björn et al., "The Impact of Anthropomorphic Interfaces on Influence, Understanding, and Credibility," Proceedings of the 32nd Hawaii International Conference on Computer and Systems Sciences, Maui, Hawaii, 1999.

Nass, Clifford, Jonathan S. Steuer & Ellen R. Tauber, "Computers Are Social Actors," Conference Proceedings on Human Factors in Computing Systems (Boston, MA: 1994): 72-78.

Piaget, Jean, The Child's Conception of the World (London: Routledge & Kegan Paul, 1929).

Reeves, Byron & Clifford Nass, The Media Equation: How People Treat Computers, Television, and New Media Like Real People and Places (Stanford, CA: Cambridge University Press, 1996).

Shaym, Sundar S. & Clifford Nass, "Source Orientation in Human-Computer Interaction," Communication Research 27:6 (December 2000): 683.

Mikael Jakobsson:
Mikael Jakobsson is working on a Ph.D. on social interaction in virtual worlds at the Department of Informatics, Umea University. He is a part of the Net-Life Research Group where he edits the Net-Life Resources. He is also part of the Creative Environments research group at the School of Art and Communication, Malmö University.  <mjson@informatik.umu.se>

©1996-2007 RCCS         ONLINE SINCE: 1996         SITE LAST UPDATED: 12.10.2009