An Essay Evolves / Making models
  • If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • Social distancing? Try a better way to work remotely on your online files. Dokkio, a new product from PBworks, can help your team find, organize, and collaborate on your Drive, Gmail, Dropbox, Box, and Slack files. Sign up for free.


Making models

Page history last edited by PBworks 12 years, 12 months ago

Creating conceptual models


Why am I doing this? Find out more on my blog.


It seems to me that this is quite a tough thing to do. It also seems that if I am going to critique someone else's model-making, I should have a bash at it for myself now, before I read too much. I'm going to propose a model of memory here, and I'd like as many of you as possible to critique it. I'd also like you to post your own. I'd especially like the non-psychologists to have a go. And perhaps if we can come up with a really good one I'll work it into the final draft of the essay.


Here goes. My model is based on certain assumptions, as discussed in First thoughts on memory. These are that memory is modular and that there is a long term, short term and sensory aspect to it. I haven't thought it through very much, I just want to get it out there.


There are three horizontal layers in my model, and these are fragile to a greater or lesser extent. Perhaps they're gelatine sheets of varying thicknesses. The uppermost is really very fragile, the middle one much less so and the bottom one is quite resilient, so perhaps it's not gelatine but a kind of rubber which is quite tough but vulnerable to perishing with time and pressure. So, experience rains down upon the sheets in showers which are sometimes very light, sometimes moderately heavy and occasionally like a tropical rainstorm. 


The first layer is like our senses - everything falls on it. Sometimes, if the fall of experience is very heavy in places, it gets through to fall on the second, slightly tougher layer (which is the STM). Particularly heavy experience leaks through the STM layer to the one below (the LTM), or sometimes we poke holes in it to allow the drops of experience through (this is analogous to when we actively use memory techniques). Below, the LTM layer sags under the weight of our experience. Occasionally it leaks or perishes (the effects of illness, age, and maybe just of having too much stuff in it).


Sometimes, experience isn't heavy enough or frequent enough to get through a particular layer. But where it has impinged upon the gelatine, it has left its impression. So, if the drops fall in the same place, there is less resistance. This is like when we acquire a particular skill, don't do it for a while, but then happily find that we re-learn faster than a total newbie.


Ok guys, what do you think? How could I test this experimentally? Where does it have room for improvement? What can you suggest (especially all you design students)? There are no rules, so make free.


Playing devil's advocate


Right. I've been thinking (rumble click squeal). I have a duty to try and imagine what a system of memory which was not modular might be like. This might start with the idea that perhaps we don't forget things in the sense that they are erased from our neural architecture. It might be that all of our experience irreversibly affects us, that we actually 'remember' everything, whether or not we can get conscious access to it. Perhaps errors in memory are somehow down to problems with the attentional or even emotional aspects of remembering. 


There exists some famous evidence that we can easily be influenced to 'remember' experiences we have never had. This tells us that memory isn't a matter of file-and-retrieve, but of something dynamic - a process of reconstruction. Perhaps it depends somehow on a sensory experience having an accompanying feeling of either 'nowness', 'thenness' or 'way back whenness'. So, if we rehearse a phone number we wish to remember, repeating it many times, the passage of many phone number events will cause us to have a feeling of 'way back when', an impression that many chunks of time have passed. If we invest a lot of attention in something, we get a sense of 'way back when'. Likewise, if something has a strong emotional overlay, we have the same (possibly because it commands our attention). Neutral events are tough to pair up with any emotion or attention (think about the task of remembering a page of your local phone directory - tough, huh?). This is all about remembering an interpretation, I guess.


If we buy this idea, all memory is working memory.


That's as far as I can go. Hey, thanks for all of your ideas. As well as posting them as comments, please feel free to edit this page or even to make a page.


I think there are two things to be considered and addressed with the 'holes' and 'gelatin' model... the first is a kind of neuropsychology argument which says (a) that anything novel will get our RAS responding and that assists us to direct attention to whatever it is that is being presented to us for evaluation of some kind - so something significant enough e.g. a scary/painful event, ('one-trial learning') doesn't need repeated exposure. The second argument comes from psychodynamics which would argue that if our evaluations (made at preconscious level) of any phenomenon are negative then they don't get through into consciousness.


Interestingly enough the work on ultra-short-erm memory phenomena suggests that there are (at least) 2 levels of attention and that conscious awareness (and therefore memory processes) begins with Attention 2. Attention 1 is almost like a preconscious level where phenomena are picked up by the senses and available to us to report on but they are not transferred into conscious awareness. They have something of the 'subliminal' about them (but that's a whole other field of memory research) and to us they seem like 'wild guesses' or 'hunches' when we pick up on them - Freud's 'TOTP', 'Tip-of-the-Tongue' phenomenon has similar roots. Again, in the professional development field Michael Eraut talks about how another kind of 'out-of-awareness' memories can still be utilised by people through what he calls 'Tacit Knowledge' - the skilled people whose 'hunches' are so often right....


The concept of memory is an interesting one - yet what is memory? do we differentiate between short and long term? Is memory differnet at differnet times in our lives - for example the items that we store and the way that we store them differ - tieing in to Abraham maslow and the hirechy of needs. Sleep and dream memory and elements that rarely are considered and yet what role do they play in our lives? Thinking outside of the box it is arguable that some definitions of memory can equally be applied to machines - cookies, log in information etc.





Comments (5)

Anonymous said

at 12:08 pm on Feb 13, 2007

I like your rainstorm analogy! But what about the odd things that aren't "heavy rain" but stay in your LTM anyway? Like a pair of shoes you had as a little girl? Or what a friend was wearing in the school photo when you were 6? Would these things be heavy rain without us realising it or would there be holes for these memories to get through created by something else?

Just a thought, I'll stop waffling now!

Anonymous said

at 12:25 pm on Feb 13, 2007

Yes, yes! Excellent point. Some things do just stick without effort, don't they? I believe in a dynamic unconscious, so off the cuff I would be more likely to say that this kind of stickability is a property of the system rather than the rain - maybe areas of pre-weakening formed by an accumulation of related micro-experiences? Hmm. Food for thought.

Anonymous said

at 12:37 pm on Feb 13, 2007

I'm a big fan of analogies.

What if memory is like sedimentary rock formations (layers upon layers built up over time)? The exposed surface constantly has 'bits' fall on top of it. Some 'bits' are immediately weathered away while others remain. Of those that remain, more 'bits' cover them rendering them no longer on the exposed top-most layer. The deepest layers (representing LTM), despite being far removed from activity on the exposed surface, are, nonetheless, subject to fault lines, pressure, heat etc. The location and composition of the deeper layers can therefore determine the longevity (or lack of it) of long-term memories.

I guess this analogy implies the distinction between LTM and STM as being of degree rather than kind. Whether this is accurate I wouldn't presume to speculate.

Anonymous said

at 5:44 pm on Feb 13, 2007

To add to the analogies you could think about memory in financial terms. Working memory would be a steady income, STM a slush fund to dip into as and when required and LTM a speculative investment that may or may not bring bring future returns.

Anonymous said

at 10:15 am on Feb 15, 2007

I think that the various models and modes of memory must be heavily influenced by our personality type also. for example - a shy person lacking in self esteem might go over and ver that 'humiliating' class room experince (of getting the answer wrong, of mispromouncing something ....) - and build up a ral memory and experience of academic failure, say. whilst another personality type could have the same experience and literally not give it another thought - thus it is not remembered and has no long term existence as a memory and no long term impact on the thoughts, feelings and behaviour of the person involved. Pehaps there is a link with notions of self-efficacy (Bandura)?

You don't have permission to comment on this page.