August 11, 2014

Memory Distortion vs Efficiency in Remembering

Causality provides mnemonic ordering within a narrative sequence. Where a story reduces the experience of real life to a manageable collection of "events", causality keeps those story elements connected and sequenced. By definition, causality distorts actual causation, just as maps distort geographical features and stories distort lived experience (personal, vicarious, or imaginary) in their respective domains of figural representation. But since the particular advantage provided by causality is a mnemonic efficiency, it may be that causality itself should be characterized as an instance of memory distortion.

The above paragraph summarizes my recent blog series and explains why I must now dive briefly into the category of memory distortion, per se. So, without further ado...


The study of memory distortion is currently centered within cognitive psychology, a field I have barely begun to scratch personally, but Anthony Le Donne (in The Historiographical Jesus, 2009) provides a helpful summary of work by Michael Schudson, who offers four categories of memory distortion, two of  which bear specific mention here, briefly. [Note: Schudson's work was included in an interdisciplinary study by prominent research psychologist Daniel Schacter; Schudson himself taught graduate level journalism at Columbia with a PhD in Sociology.] The first category is called "truncation" or "distanciation", in which memories grow vague and lose detail. The fourth category is "narrativization" in which memories become "conventionalized through the constraints of storytelling. Without space to review the helpful aspects of both these scholars' valuable work, I will merely note that "truncation" seems to fit my own thinking about the basic "reduction" from experience to story, and "narrativization" seems primarily oriented around what Hayden White called "emplotment", or what Seymour Chatman called "discourse". In fact, Le Donne discusses "truncation" (p.62) as the narrative curtailing of time by a storyteller's selection of "beginning", "middle" and "end", and Le Donne cites Hayden White multiple times in discussing the distortion category of "narrativization".

These are valuable and practical observations of storying memories - or, are they valuable and theoretical observations of memorializing particular narratives? - and they will remain so even if Schudson's categories could be further scrutinized in comparison with narratological theory. But for that matter, as things currently stand within the larger field of cog-psych, the suggested taxonomies of memory type and distortion type seem to proliferate easily, and perhaps endlessly. For instance, Schudson's suggested categories appear in an edited volume by Daniel Schacter, an eminent psychologist and prolific researcher of memory distortion. But Schacter's edited volume alone (Memory Distortion, 1995), contains a daunting volume and (to me, at least) a bewildering array of research, theories, and categorizations, based on the contributions of over two dozen scholars. There is obviously some refining work yet to be done in this field.

One pleasant discovery, just two weeks ago, was the following statement in the abstract of a recent paper by Schacter (et al) on Memory Distortion (2011): (emphasis mine)
Here we integrate emerging evidence that several types of memory distortions - imagination inflation, gist-based and associative memory errors, and post-event misinformation - reflect adaptive cognitive processes that contribute to the efficient functioning of memory, but produce distortions as a consequence of doing so.
While I can't begin to explain what these listed "types of memory distortions" might entail, I was happily flabbergasted to find these words: "processes that contribute to the efficient functioning of memory, but produce distortions as a consequence of doing so". Even though I'm not entirely sure what Schacter means by "efficiency" - his recent paper remains less than clear to me until I learn the referential meanings of cutting edge cog-psych jargon; but thank you, God, for that abstract! - or whether Schacter's "efficiency" is anything like the kind of "efficiency" I've been thinking about since March and blogging about since June, this has been an encouraging reason to bring cognitive psychology into my radar. I will pursue memory distortion anon.

For the moment, here are a few helpful, significant gleanings.

One of Anthony Le Donne's personal innovations is frequently to substitute the word "refraction" instead of distortion, citing the brilliant metaphor of a telescope, which refracts starlight and "distorts an imaged object in order to magnify it." He goes on, "the viewer is able to perceive an approximate distortion of distant objects not visible to the naked eye" and points out that the modified nature of a refracted image is precisely the reason astronomers find their telescopes to be so valuable. The lens refracts, but it magnifies, and thus provides images which could never be seen at all, otherwise. Drawing the metaphor back on point, Le Donne says, "memory's primary function is to render the past (which is invisible to the naked eye) intelligible to the present" even if this requires an "acceptable approximation".

As I said in part 1 of my recent blog series, Anthony's work has been a primary inspiration for my re-thinking of narrative function. Add to that my own peculiar passion for chronological study and I found myself wondering how "Memory" might encapsulate "Time", as in "Time itself". That led to my recent series of posts, in which I've been discussing how Story and Causality increase the efficiency of Memory, specifically how consequence embeds sequence, allowing the mnemonic ordering of time, as I said at the top.

In that series, I have paused just after promising to illustrate more thoroughly how this "mnemonic efficiency" works. Before I could do that, I felt it was important to compare my own thoughts on "efficiency" with the literature on "distortion" (at least, as per my budding and brief forays into cognitive psychology, as indicated above). In concluding this post, therefore, here is what I feel tentatively confident enough to declare, regarding my own developing project.


First, by definition, causality is a representational distortion of actual causation. Regardless of whether causality distorts due to the imperfections of human perception, or the authorial bias of narrativization, or the cognitive processes of psychological "distortions", the distortion of causality is primarily akin to the distortions of other representations, such as maps and narratives and telescopes. (Though Le Donne applied his metaphor to memory, proper, the technology itself is a tool of representation, as are binoculars, cameras, some convex mirrors, eyeglasses, and perhaps even the human eyeball itself, all of which distort the appearance of objects in order to make them more visible, or indirectly visible despite handicap or obstruction.)

Second, causality is a distortion with a specifically mnemonic advantage, in that it more easily accommodates the remembering of sequential events. This is not like Schudson's categories in which (per Le Donne) distortion is what enables memory to exist at all in the first place, because "Memory is distortion." (LD p.51) or as Le Donne quotes Jan Assmann as saying: "The notion of 'distorted memory' seems to presuppose that there is something like "undistorted memory." As it happens, I completely agree with these points. Nevertheless, the advantage of causality (as I tend to see it) is not something which allows memory to exist. Rather, causality allows a single memory to encompass more data.

Third, this accomodated efficiency of causality is, itself, an enhancement of what I've described as the efficient accomodation provided to memory by "story". However, more in line with the previous point, the phenomenon of "story" may itself be classifiable as a memory distortion, in that "story" allows memory to exist in the first place. Again, the difficulties here are more taxonomic than pragmatic, but we do need to find ways of discussing such things.

((***For instance, there arise at this point further questions about Schudson's first and fourth categories (Cf. Le Donne's citation of Fentress and Wickham on story and "plot", LD p.55) which closely mirrors the subjectivity of dividing "story/plot" from "discourse/narrativization". If memory theory is brought more in line with narratology, these categories of distortion may prove as incrementally subjective as the division between "plot and emplotment", or "story and discourse".  (See my Memory & Narrative, parts 1-6, and especially my "Story is Subjective, Temporality is not".) In trying to identify the plot points of a narrative sketch, it almost seems like we might as well work towards a calculus of narrativity, but there is no "instantaneous rate of change" in a narrative arc, because all the points we might "plot" are entirely subjective. On the other hand, by that precise logic, there might well be a point of change to be seen, by some level of perception, at ANY particular point in a narrative arc. Thus, again, Schudson's first and forth categories can blur just as much as Chatman's categories. However, in the end, these two categories - "truncation" (what I would rephrase "reduction") and "narrativization" are every bit as useful as "story" and "discourse" when viewed more as rough approximations than precise categorizations. One wonders if all taxonomists in cognitive psychology will be willing to hold loosely to categorizations without letting go of their efforts to understand and to illustrate memory distortion. ***))

Fourth, as causality increases the efficiency of storied remembering, so there are degrees of perceiving causality which can further enhance both the strength and efficiency of remembering extensive sequences. What this means, in simpler terms, is that mnemonic efficiency increases with the strength of causality.

Fifth, and finally, it is worth noting again the severe dubiousness of dividing between "story" and "discourse" as a way of observing that causality can arguably be identified with both categories. On the one hand, given its role as the central aspect of plot, causality can be aligned most closely with "story" (or Schudson's first category). On the other hand, given that the prioritization of causes (the relationship between "necessary" and "sufficient" causality) is in large part what's generally being provided by narrativization (aka emplotment, aka "discourse"), causality can also be aligned very closely with "discourse" (or Schudson's fourth category).

Ultimately, therefore, it seems causality can be defined either as a category of memory distortion - in that a Plot is the basic form of a Story, which is one kind of representational distortion that allows Memory to exist - or as a mere enhancement of mnemonic efficiency, as I've been describing it in my previous work.

If there is to be any definitive taxonomy here, I will leave it to much greater minds than my own.


What seems most workable for my own purposes, therefore, is to persist with the metaphors of distortion but focus on aspects of story and narrative which increase the efficiency of remembering. This emphasis cuts straight to the heart of my ongoing project. As I said in Memory & Narrative, 2:
Which comes first? Do we construct a story with immediate gestalt perceptivity and then work to remember it? Or do we form short-term memories of experience and then construct a story selectively from among that material?
My working concept begins with the process of daily, tacit autobiography. We experience things, forming short term memories, most of which fade quickly. The outstanding patterns of memory persist most vividly. Then we narrativize, or more properly speaking, we "story-ize" from those memories. We reduce, we conflate, we amalgamate, we truncate, we summarize, and we eventually narrativize. (Perhaps we narrativize slightly after we story-ize, albeit not long after, or infinitesimally "not long after", in neurological time). As time continues to pass, we repeatedly re-narrativize, but the initial "storying" of immediate memory is what tends to cement the most lastlingly. This may indeed corresponds with the cognitive study of forgetting in psychology, in which memory content is reduced geometrically, in proportions resembling the graph of a logarithm function. But on that note, I'm once again slightly over my head.

But, dear reader, you've got all that, right?

In short, I believe we need to start looking at memory and story within multiple layers. We remember. Then we story. Then we memorialize the story. Then we re-narrativize that memory. And so on, ad infinitium.

But anyway.


What is the relationship between "Memory Distortion" vs "Efficiency in Remembering"?

For now I will continue to write about story and discourse as being ways to increase memorability that also cause some distortion.

My next post will be part 7 of the ongoing series, in which I plan to illustrate ways that popular views of the past mimic memory distortion in the ways they generate efficiency in remembering stories.

See you then...

No comments:

Recent Posts