March 11, 2010
People Consider More Possibilities For Past Than Future?

Some Brown University researchers think humans consider more possibilities when diagnosing than when predicting.

Fernbach and the other researchers explored the degree to which people are overly focused on a single cause when pursuing two fundamental kinds of thinking predicting the likelihood of an outcome and diagnosing the causes of an outcome.

They see these two kinds of thinking as flip sides of the same coin. Predicting outcomes calls for thinking forward from the cause of the outcome, such as predicting the likelihood that someone who goes on a diet will lose weight. But offering a diagnosis involves thinking backward from an outcome to the cause, such as diagnosing whether someone who lost weight dieted.

The researchers conducted three studies with medical professionals and Brown undergraduates. Their findings: In each case, the subjects considered alternative causes when they made diagnoses, but did not do so when making predictions.

I can see a way to try to use this result to think more productively: When trying to predict the future list some possible paths. Then for each path imagine you are in the future and a series of events caused developments to happen along that path. Think back from this imagined future vantage point and try to identify the causes of the outcome. If you do that for each possible outcome you might shift your mind into a more backward-looking diagnostic mode.

Share |      Randall Parker, 2010 March 11 11:13 PM  Brain Cognitive Flaws

phos said at March 12, 2010 7:38 AM:

That's why advertisements have before and after photos- NOT after and before photos.

If you saw the result first, you might consider alternative causes.

James Bowery said at March 12, 2010 7:38 AM:

From an informational standpoint, this makes no sense. In predicting the future, you have the present. In predicting the past -- "diagnosing" -- you have not only the present, but memories of the past. Quite simply, the past is more informed than is the future so the probability distributions are more sharp hence relatively plausible possibilities more limited.

Lono said at March 12, 2010 8:48 AM:


This is true - but I think the point was the subjects had no historical information on the scenarios presented - so they were just speculating on the past - as they could with the future.

The surprising thing - which isn't really surprising exactly for the reasons you presented - is that people normally have more information on past events and thus can make a more accurate diagnosis than a future prediction.


I came across some New Years Resolutions type spreadsheet I must have made shortly after leaving University - and I was a little concerned how few of those "goals" I had kept/achieved.

Certainly it was an ludicrously aggresive and overly opptimistic list in hindsight - and some of my professional interests changed - making some of my resolutions unecessary - but my mild sense of regret on some of the unaccomplished items made me think in a manner very similar to the one you proposed.

In fact I made a new Resolution list and then "role-played" the idea that I had not succeeded in several key ones - I then tried to simulate diagnosing why I had not achieved these key "future" goals in the period I expected - and I must admit I did have much more clarity then if I had been blindly attempting to predict potential obstacles otherwise.

So - from my own experience - I do believe your thought exercise has great merit for those who seek to improve themselves - or to become more efficient in some way.

I think going forward I will try to do this at least a few times a year - as I have found with older age the perception of time is greatly reduced - and complacency is a far more difficult trap to thus avoid.

(why perception of time changes with age is an extremely interesting aspect of the Human Condition - one that deals with the varying percentages of white versus grey matter in the brain - a subject your blog - and Scientific American has covered previously in excellent detail)

PacRim Jim said at March 12, 2010 6:03 PM:

The past is ever so much more malleable than the future.

Mirco said at March 14, 2010 2:38 PM:

Modeling the causes from the present to the past is standard procedure in economics. Every step enlarge the numer of causes, but the fact that the present conditions must be true limit the number of possible past conditions. The reverse is not true.
Thus is because setting a desirable (or undesirable) future condition and go back to the causes that would cause it is the best way to predict the future.

David Gobel said at March 16, 2010 5:48 AM:

envisioning the future as fact and working backwards is how I think. For instance, Every 5 years I write a "last will and testament" in favor of my future self 5 years hence, envisioning what that newer version of myself would need and/or appreciate having, being and avoiding, and then working backwards from that future "unsealing of the will" and pruning the behavioral/event tree of the intervening years so as to proceed successfully toward that future. The number of inheritances are limited to no more than 3 - 5 elements/assets to make them practicable/tractable/visionable.

current goals for 2015,

decouple from currently non-optional resource drains (public utilities for instance)
be biologically and emotionally younger than today
have the strong ability to generate positive good for self, family and all
out-compete competitor species (corporations, institutions, non-biologics)

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright