The Homunculus Effect
Talking about Memory
Reality Monitoring
Excerpts from Marcia K. Johnson
(Revised September 21, 1997)

Johnson, Marcia K. (1991). Reality Monitoring: Evidence from Confabulation in Organic Brain Disease Patients. In Awareness of Deficit After Brain Injury. Clinical and Theoretical Issues. Edited by George P. Prigatano and Daniel L. Schacter. New York: Oxford University Press, 176-197.

"Because humans have a cognitive system that takes in information from a number of perceptual sources and that can itself internally generate information as well, one of the mind's most critical cognitive functions is discriminating the origin of information. We constantly use this ability in considering ongoing experience (Is what I see now 'out there,' or am I only imagining it?) and the products of past experience (Is my memory for an event that happened when I was 5 years old a memory for an actual event or an event I imagined as a child?) (180).

"I have suggested we use the term reality testing for the processes by which people make such distinctions during ongoing experience and the term reality monitoring for the processes by which people discriminate between memories derived from perception and those that were reflectively generated via thought, imagination, dreams, and fantasy" (180).

"Reality monitoring failures occur when people confuse the origin of information, misattributing something that was reflectively generated to perception or vice versa. That is, reality is not directly given in perception or remembering but is an attribution that is the outcome of judgment processes" (180).

Memories originating in perception typically have more perceptual detail, while memories originating in thought typically have more accessible information about cognitive operations. "Differences between externally and internally derived memories in average value along these dimensions or attributes form one basis for deciding the origin of memory" ( 181). She calls this first type R-1.

"A second type of decison process is based on reasoning"--for instance, you might have a memory of flying, and realize that this is not possible; or of having visited Shanghai, when you know you never have (my examples). You may also assume that "someting that comes to mind quickly is likely to be an accurate memory of an actual event" (181). The calls this second type R-2.

"Thus, reality monitoring most likely produces errors when perceive and imagined events are similar along dimensions that normally provide a discriminative cue (e.g., if the imaginations in question are particularly rich in perceptual and contextual detail), when reasoning fails, whe the relevant background knowledge is not retrieved or unknown, or when metamemory assumptions are inaccurate" (181).

"if perceptual qualities of imagined events were unusually vivid, they would be more difficult to discriminate from perceived events. It might happen, for instance, if reflective processes were especially successful in recruiting perceptual processes during imagination, as is evidently the case with good imagers" (183).

"the vividness of a memory may indicate that the event actually happened," while an assessment of "its plausibility may indicate that it could not have happened" (190).

"Reality monitoring is a fundamental memory function that anchors us in a perceived external world, in a felt past life with an autobiographical quality, and in a network of knowledge and beliefs that we take to be derived from experience in a veridical way. At the same time, reality monitoring produces in us a compelling sense of ownership of our own ideas, fantasies, and hopes. It is only when the boundary between externally derived and internally generated information becomes blurred, as in the case of confabulation or delusions, that we can fully appreciate how central this discrimination is to defnining the characteristics of normal mental experience and to functioning effectively in the world."

Johnson, Marcia K., John Kounios, and John A. Reeder (1994). Time-course studies of reality monitoring and recognition. Journal of Experimental Psychology: Learning, Memory, and Cognition 20: 1409-1419.

Johnson's model (outlined above) "suggests that people do not remember the source of a memory per se; rather, they infer the source of a memory…. Instead, they make inferences or attributionis based on available cues in the form of various types of memorial information" (1409).

Now, "if reality-monitoring processes are based on heuristics rather than on direct dectection, they should be systematically error-prone" (1409).

"Memories for imagined events would be expected to include more available information about the cognitive operations that established them, and memories for perceived events would be expected to include more perceptual detail" (1416).


Johnson, Marcia K. (1991).Reflection, Reality Monitoring, and the Self. In Mental Imagery. Ed. Robert G. Kunzendorf. New York: Plenum Press, 3-16.

"A self," Johnson writes, "is a byproduct of reality monitoring processes that distinguish perceptually-derived from reflectively-generated information." She argues that the self-as-source forms a locus for the attribution of the reflectively generated information, while the phenomenology of the self-as-controller arises in various interactions of the reality monitoring system, notably between the perceptual and logical levels (13).
 For further references, see Reality Monitoring in CogWeb's Bibliography.

The Homunculus Effect
Next: Talking about Memory
Talking about Memory
Return to the CogSci index
Previous: The Homunculus Effect
1997 Francis F. Steen, Communication Studies, University of California, Los Angeles