I breezed through Jim Benson's short and informative Why Plans Fail: Cognitive Bias, Decision Making, and Your Business. As you can see from the subtitle, it isn't about blaming someone else for why plans fail. It's about helping us see how our own thinking gets us in this mess.
I have seen many discussions of cognitive biases in popular writing (Gladwell, the Heaths, etc), but those books tend to be much more discursive. I like that Benson has decided to keep the discussion closely limited to cognitive biases connected to planning and managing knowledge work. And that he is drawing from his own experience and that of his client engagements.
I could comment on nearly all of the biases and thinking errors that Benson describes. They are all so evident in things I do every day - and things I see in my interactions with others. His book doesn't try to talk about how to overcome these biases, other than suggesting that there is power in knowing that they exist. That said, knowing about these biases is a starting point to thinking about how to eliminate the effect.
The best example of this is his discussion of the planning fallacy "and its helper apps." Our estimates (predictions, really) are taken as gospel and then we are surprised when they don't come true. Given that, maybe we should be doing a lot less estimating and much more doing. Plan a general direction, move along that path, and then check to see if the wind hasn't blown you off course. Then correct. Plan-Do-Check-Act. Observe-Orient-Decide-Act. Take your pick. But act. Don't drown in trying to be perfect, just get up and get going WITH a commitment to stop and check that you are still heading in the right direction.
Of course, this leads me to my own conflict. I am very comfortable with the idea that we can't predict how long project work will take. It's a core tenant of the Critical Chain Project Management work I've done over 10+ years. When dumping these estimates into plans, the claim goes that people put in all sorts of padding to account for interruptions, multitasking, shifting priorities, Murphy's Law, etc, etc. An element of the CCPM solution is to reduce the padding and create a time buffer for the whole project, so that when anomalies happen, the buffer can absorb the change and the project remains on track.
The problem I find here is that Benson references Hofstadter's Law that says people consistently under-estimate the time it will take them to do something, even if they know they are going to underestimate. If they are underestimating what does this mean for the CCPM crowd? Is there less padding than we think? But the overwhelming evidence from my experience and that of others in CCPM is that project networks are built with significant "extra time" in the individual tasks. I think this may be the source of comments I have seen that suggest project management is a symptom of a larger problem - rather than a solution.
If you are curious, there are over 100 types of cognitive bias listed on Wikipedia. Not all of them are connected to planning - though many could be.