PsyFi Search

Wednesday 26 January 2011

Financial Memory Syndrome

False Memory Syndrome

There continues to be serious debate over the concept of false memory syndrome, where allegedly innocent people have been accused of serious crimes on the basis of memories which may have little basis in reality. Whatever really lies behind these cases it seems that planting fake memories in people’s brains is rather too easy to make the truth easy to ascertain.

The problem is that our memories are rather more malleable than we’re brought up to believe and it’s easy to create imaginary ones if you know what you’re doing – and often if you don't. Even worse, we can plant fake memories in our own minds and these are implicated in many of the behavioural issues that dog investors. Financial memory syndrome is at the heart of many a financial crisis.

Memory Management

Slowly and painfully psychologists have teased apart our various types of memory. At the highest level there’s short-term and long-term, where the former is the scratchpad that allows us to remember telephone numbers long enough to dial them. Seven plus or minus two is the rule of thumb for the number of “chunks” of information we can remember transiently, although age, alcohol and political interns can reduce this significantly.

Short-term memory isn’t implicated in most false financial memory scenarios, although as Buffett and Munger have wryly pointed out, ideally you want to invest in companies so simple they can be run by someone with Alzheimer’s disease. They ought to know, because they did with Louis Vincenti at Westco

Lost in a Mall

Memory researchers have gradually teased out some rather scary results, if you’re of the view that you’re in control of yourself. In a particularly famous experiment, described in this paper by two of the authors, Elizabeth Loftus and Jacqueline Pickrell, it was shown how relatively easy it is to plant false memories by extending a range of existing evidence. The researchers created a situation in which a 14 year old boy was induced to believe they’d experienced an entirely fabricated story of being lost in a mall: it’s not just the experience which ends up being recalled but a whole host of extended details which aren’t even part of the experiment:
"When asked to describe his memory of getting lost Chris provided rich details about the toy store where he got lost and his thoughts at the time”.
When told one of the memories he’d been presented was false:
“He selected one of the real memories. When told that the memory of being lost was the false one he had trouble believing it”.
Misremembrance of Things Past

The malleability of memory appears to come out of the mechanics behind it. We simply can’t remember all of the details of everything that happens to us so we resort to short-cuts. In particular memory fades quite quickly and when asked to retrieve them we seem to merge a general template of similar situations – a so-called schema – with the few details we can retrieve to complete a “whole” memory. Unfortunately our remembrances of things past are far from accurate.

Additionally our reliability as witnesses to our own lives is often somewhat undermined by the fact that, as the psychologist Richard Wiseman has shown, we often fail to spot the strangest things if our attention is focused elsewhere. In order to record stuff in memory we need to notice it in the first place (make sure you follow the instructions).

Human memory is fallible and malleable and we don’t even know it. Memory research shows quite clearly that eye witness testimony is unreliable and should never be relied upon unless corroborated in other ways. Yet in court cases we rate such evidence as overwhelmingly more important than other sources, presumably because we believe our own memories are perfect rather than reconstructed palimpsests of dubious authenticity.

Behavioral Memory

Many behavioural biases are implicitly linked to memory failures. Hindsight bias, for instance, in which we believe we previously predicted stuff that we didn’t and from this draw the erroneous conclusion that we can predict the future is clearly associated with our inability to correctly remember our own expectations. Our habitual overconfidence couldn’t be justified if this wasn’t the case.

Generally, then, memories decay quite quickly which is likely to be why we’re so myopic about tail-end risks: they simply don’t occur often enough for us to encode them as significant events and unless we take the effort needed to learn the lessons of history we simply surf our way from crisis to crisis like goldfish: every crash a unique experience, the same piece of seaweed a continual surprise.

Memory Trauma

Just to round this square off nicely, we also find that traumatic events get burned into memory so effectively that we never recover from them. As Malmendier and Nagel report about people brought up in the Great Depression:
“We find that individuals who have experienced low stock-market returns throughout their lives so far report lower willingness to take financial risk, are less likely to participate in the stock market, invest a lower fraction of their liquid assets in stocks if they participate and are more pessimistic about stock returns”.
Perhaps the more general finding is that people expect the conditions they learn to invest in to continue. Sadly the lessons of the past are that we can usually expect that they won’t or, even if they do, that they’ll be interrupted by the periodic equivalent of asteroid strikes.

Memory Flaws

In fact memory research has come a long way over the past sixty years and even a brief summary can’t do justice to the range of strange effects that can occur and the ways in which they can impact investing behaviour. Fortunately Joackim Klement has published a paper on The Flaws in Our Financial Memory, which describes them in far more detail than possible here.

In summary, though, the paper draws out a number of failures in long-term memory which can damage our investing returns. Transience, for example, is essentially the problem that our memories fade without constant refreshing and is likely implicated in recency and primacy effects, where we invest in that which we most easily bring to mind. An even better one is cryptomnesia where we attribute other people’s ideas to ourselves, because we can’t actually remember where we got them from.

Memory Anchors

Another problem area is anchoring where certain words can be used to frame our ideas. This eems to be bound up with and closely associated with suggestibility. As Klement states:
“The performance of the S&P500 in the year 2008 was -35.8% yet respondents will likely provide different estimates of the annual performance in the following three cases:
  1. What was the performance of the S&P500 in the year 2008?
  2. What was the performance of the S&P500 during the bear market of 2008?
  3. What was the performance of the S&P500 during the crash of 2008?”
In reconstructing memories we’re wholly at the mercy of our behavioural biases. We can be induced to unknowingly follow the suggestions of others and we’ll typically forget stuff that we don’t like and recreate it in our own self-image. When it comes to financial matters we find it ever so difficult to recall our failures and so much easier to remember our successes. It’s yet another reason why keeping a diary of our investing activities is so important, there’s no other way we can even half-way guarantee a correct analysis of our past actions.

Rollercoasters

The flaws in human memory are many and manifest and make it an extremely unreliable source of information for data about our individual experiences in markets. This isn’t a general observation about the limitations of peoples’ investing abilities – it’s a specific observation that applies to each and every one of us. It’s testimony to the power of humanity to delude itself and the power of group processes that we continue to operate on the basis that markets are a one-way ticket for ourselves and a rollercoaster ride for everyone else.

Of course, for those of us that choose to forget the lessons of the past the markets are a one-way ticket. Just don’t plan on retiring any time this century.


Related articles: Investors, You've Been Framed, Hindsight Bias, Overconfidence and Over Optimism

3 comments:

  1. Your observation on hindsight bias and overconfidence explains why investors and (particularly) financial advisors think they routinely outperform the market even if there is no evidence to support it. We remember our winners and (conveniently) forget our losers. We also remember the many times we correctly forecast the future but forget the times the forecast failed to happen. My grad school professor Meir Statman framed it this way: You dreamed about Aunt Minnie the night before she passed away and now think you are psychic. You forget about all the times you dreamed of Aunt Minnie and nothing happened.

    ReplyDelete
  2. Funny, I thought it was all implanted in that ball they shoved up my nose!

    Sometimes we use memory to hold ourselves responsible for mistakes that weren't really mistakes. For example "buyer's remorse" or "what was I thinking?" What we are remembering was only the fateful decision that led to the outcome that we now recognize to be bad, and not the circumstances around it. Instead we backtrack today's context into the past, so of course our thought process looks silly. Probably a stress factor at dinner tables in many underwater equity households today. I've rented all my life but I don't believe recent homebuyers did anything stupid or wrong.

    When the consequences are life-altering, the decision process gets ingrained, probably some sort of evolutionary brain shortcircuit that prevents us from repeating something that almost killed us. What I started trading stocks, my dad expressed concern about my trade size. So I asked him "what's the worst that could happen? If I lose it all, I can afford it". His reply was "no the worst outcome would be if your first bet won big, like a 100-bagger. Then you'd be screwed for life." Luckily for me my first several punts paid off only a few percent that first year.

    ReplyDelete
  3. Hi Kent

    There's research going back as far as 1915 when a guy going by the name of Dan Guyon showed that investors lost money even while the markets were going up 65% - and thought they'd done well. He who forgets history is destined to repeat it ...

    Always the same lesson: keep a diary :-/

    ReplyDelete