# Can attention be improved to reduce the frequency of mistakes?

There is a class of error that is sometimes colloquially referred to as "not paying attention". These are the sorts of error where the subject knows the correct way to perform a task, but fails to follow it due to a lapse in whatever mental process is responsible for ensuring that actions are carried out according to intention.

Some examples are:

• A mathematically competent adult solving an advanced math problem incorrectly because of a simple arithmetic mistake, such as 5+3=15.
• A well-prepared student getting a multiple choice question wrong because it was asking "which of the above are not true", but she mistakenly selected the ones that are true.
• A skilled roboticist damaging an expensive circuit because he accidentally wired the components incorrectly.
• Mixing up two terms which refer to different things, despite understanding very well the concept that either term refers to.
• Typos and simple grammar errors.

Note that I do not necessarily mean attention in the sense of being able to concentrate on and pay attention to a topic. I am specifically talking about the ability to not make mistakes (where mistakes are simple errors, which the subject is capable of recognizing as errors, but happens not to in that particular instance - not errors committed because of imperfect knowledge, understanding, or reasoning). Thus, "mistakes" are:

• Not reproducible: If the subject is asked to repeat the task, they will probably do it correctly.
• Unrecognized: The subject believes that the task was completed correctly, until explicitly checked.
• Mental: The mistake is not due to failures in fine motor control but has to do with the output of the mind. The mistake may be due to incorrect interpretation of sensory stimulus, but not due to an overt corruption of the stimulus itself (a person with good vision "misreading a word" in a brightly lit room from a black on white page with a suitably large font counts, but trying to read by starlight on a moonless light doesn't).

So, my question: Is there scientific research dealing with "attention" in this sense (ability to make few mistakes)? What factors influence it? Can it be trained, or is it an innate, invariant quality of a person that can only be accepted and accommodated?

Also, I am asking specifically about "healthy and normal adults", not disorders such as ADHD (unless the pathology helps elucidate the healthy case). As the saying goes, to err is human - but some humans err more than others, and I am interested in understanding why (and what strategies are available to reduce this tendency).

(edited from my original question on Academia.SE)

• Welcome to CogSci! I would just like to thank you for this question, as it is a prime example of a non-scientific question which is very clear and interesting nonetheless! – Steven Jeuris Mar 17 '14 at 10:09

Sounds like a class of faults resulting mainly from a common sequence: habituation, automation, assimilation, discrimination error.

1. Habituation. The tasks you describe are generally simple, ordinary, and as you say, there is a "correct" way to do them that is known. This would encourage a heuristic-based approach to problem-solving response generation. Heuristics are generally not foolproof, but they're fast and relatively effortless. Suits these "mindless" tasks well when there are lots to do that just need doing.
2. Automation. The easier and more routine a task becomes is the more one will handle it without conscious, deliberate thought: i.e., without attention to detail or inhibition of first-impression-based impulses. One focuses instead on other decisions or plans; there's always the rest of the week to look forward to or worry about, and the task at hand is so boring that focusing on it is aversive.
3. Assimilation. Rote task performance that thus motivates diversion of attention for the sake of both efficiency and entertainment thereby encourages assimilation of new stimuli. In many ways, this process actually comes first, leading to habituation and automation. I bring it up now to transition my focus from the learning processes to the case-by-case response pattern that results. Assimilation is the foundation of habituation and automation, but it also results, as these are the processes of self-reinforcing feedback from assimilation decisions. More simply stated, when assimilation happens often, habits form. As habits flourish, automaticity sets in. At this point, one almost always chooses to assimilate a new scenario into existing schemas; it's harder to resist the urge.
4. Discrimination error. Motivated inattention and hastened task processing inhibits accommodation. Thus when one operates on an imperfect heuristic (that may be redundant), applying incomplete schemas to new situations that are not represented accurately in these classes of supposedly known problems leads to errors. E.g., seeing a ton of multiplication problems might bias one toward ignoring the operator, failing to notice a 45º rotation of the symmetrical cross shape, and producing 15 when presented with $3+5$. Being frustrated with my elaborate jargon and complex sentence construction might incline a reader to skim; if the reader's intended task is to edit, s/he might expect more errors where errors are to be found, and in skimming, miss the extra word in a the sentence that constitutes an error. After frequent experiences of assimilation produce automated habits, one can hardly be blamed; the tendency is innate, adaptive, and somewhat invariant in terms of simple prevalence. I.e., most normal people do it, and variation mostly exists in relative frequencies $>0$.

Excluding abnormal individuals might make the phenomenon more invariant. Abnormal function like in the cognitive tendencies of people with obsessive compulsive personality disorder might present stronger exceptions to the rule that "everybody makes dumb mistakes sometimes," even somewhat often. But then again OCPD might not be such an exception, as its theoretical representations don't entirely match its real manifestations. (Same for its fictional representations; see my answer to Why is Mr. Monk unsure?)

I'm sure plenty of research can be found on how to encourage adaptive accommodation to prevent errors. I'm inclined to look toward motivation- and personality-based approaches, as those are my specialties – hence you might say they inform my own schemas and heuristics, encouraging me to assimilate your problem as just another of the kind I'm familiar with. If I fail to respond appropriately to any subtleties in your question as a result, consider it a demonstration! I admit I've spent most of my time so far on just reformulating and recapitulating the issue you describe.

Moving briefly into some modification strategies then, consider motivating attention. This can be achieved by increasing the sense of challenge to optimal levels that induce flow, which is a state of very intense, self-motivated, even blissful focus. Just produce the challenge by requiring higher accuracy, not faster task completion.

On somewhat less modifiable notes pertaining to individual differences in error frequencies, conscientiousness and cognitive complexity seem like relevant traits. Conscientiousness relates to attention to detail and concern for error prevention. Cognitive complexity relates to adaptive thought and tolerance for complex or multitudinous schemas, which would discourage over-application of too-simple schemas for every problem encountered.