Mind wandering is
a common phenomenon in everyone’s daily lives. Thinking about lab during molecular
biology class. Thinking about this blog post during antibody incubations. Thinking
about sleep at 5:30am…okay, this might be due to fatigue. Yet, the exact
mechanisms that account for mind wandering remains inconclusive. However, there
are two prominent and competing views on mind wandering, namely the “executive
function” and the “executive failure” views of mind wandering. In this post, we
will examine both viewpoints in detail, beginning with a description of the key
components of each viewpoint, followed with a comparative analysis on the
similarities and conflicts between the two camps. Later on, in a questions and
comments section, we discuss the evidence for and challenges against each
theory. Upon consideration of both evidence and challenges, we proceed to
choose the most compelling theory. Finally, we conclude with a section on links
of these two theories to neuroscience and possible future experiments.
Summary and Comparative Analysis
Before we delve into a
discussion of the evidence for and challenges against the two models of mind
wandering, it is important to first establish what the “executive function” and
“executive failure” models entail. Thus, below, we provide a description of
both theoretical viewpoints, including key components of each theory.
Subsequently, we examine the fundamental conflict between these two viewpoints
in a comparative analysis.
Executive
Function
Conceived and detailed by Smallwood
and Schooler in 2006, the “Executive Function” model argues that mind
wandering utilizes executive resources and thus reduces the amount of executive
resources available for the primary task at hand (Smallwood & Schooler,
2006; McVay and
Kane, 2010). This view falls into the “resource theory” category in
that it espouses the view that there exists a limited pool of executive
resources responsible for information processing and attention (and lack
thereof). As Smallwood et. al. notes, “When mind wandering occurs, the
executive components of attention appear to shift away from the primary task,
leading to failures in task performance and superficial representations of the
external environment.” Importantly, the executive function viewpoint also
maintains that mind wandering “can be seen as a goal-driven process, albeit one
that is not directed toward the primary task” (Smallwood, 2006) and thus
performs an executive function.
Executive Failure
In response to Smallwood and Schooler’s
2006 study, McVay and Kane proposed an alternative model on mind wandering that
generally agrees with but extends upon Watkins’ control theory. Specifically,
they argue mind wandering is a result of “a
failure of executive control over automatically (and continuously) cued
thoughts rather than as consuming executive resources.” (McVay and Kane,
2010) in their “control failure x concerns perspective” on mind wandering. In
direct contrast to Smallwood and Schooler, McVay and Kane contends that mind
wandering “does not draw on the same executive resources as executive control,”
and thus can be controlled or prevented using the “executive–control system.”
Ultimately, this viewpoint holds that mind wandering reflects a failure in
executive control and serves no useful function.
Comparative
Analysis: The Conflict and the Conversation
I found this
week’s papers particularly interesting because they responded directly to each
other and thus allowed for a conversation on the key questions in each paper.
Thus, in this comparative analysis, we will examine the key points in this
conversation. The conflict between these
two viewpoints is clear: one theory holds mind
wandering consumes executive resources and may serve functional roles in daily
life whereas the other model contends that mind-wandering is a result of executive failure and does not consume the
same executive resources as the primary task. At first glance, this seems
like a lost case; how can you ever reconcile two completely opposite
viewpoints? Smallwood
(2010) again stepped up to the task in his “Reply to McVay and Kane.” In
response to MacVay and Kane’s contention that mind wandering is a result of
control failure, Smallwood argues that despite the possibility of mind
wandering happening after control failure, it nevertheless is “always a
conscious reportable experience and so is globally available to the system.” He
continues to argue that this global
availability suggests that mind wandering demands resources, in accordance
with his executive-function viewpoint. The most striking point in this
conversation was not the different viewpoints, but their potential
semi-coexistence. Specifically, I was surprised that Smallwood (2010) defended
his viewpoint by introducing a completely new concept, namely global availability.
What I found most compelling about Smallwood’s reply was the consistency of his
“global availability hypothesis” with the key features of McVay and Kane’s
model. However, as will be addressed in the next section, I found the concept
and introduction of the global availability hypothesis somewhat arbitrary. Even
more concerning, I found the assumptions associated with the “global workspace”
quite severe.
Questions and Comments
The Evidence
and Challenges
There exists
robust evidence both for and against these two theories on mind wandering.
Let’s begin with the Executive Function theory elaborated by Smallwood and
Schooler. If mind wandering does indeed utilize executive resources, then this
model would predict an inverse correlation between mind wandering and
performance on the primary task. This is exactly what several studies found. As
the authors note, “the evidence presented in this section suggests trade-offs
between mind wandering and task performance.” Specifically, mind wandering was
also inversely correlated with the difficulty of the primary task, which would
also support a resource theory approach. If mind wandering utilizes executive
resources, and assuming that more difficult tasks exhaust more resources, tasks
that are more difficult and thus demand more resources would leave less
available resources for mind wandering. Even more pressingly, evidence exists
to show that mind wandering actually competes for working-memory resources;
specifically, impairments (slight, but present) in task efficiency are observed
in the presence of mind wandering. The observation that mind wandering
interferes with the “successful completion of nonautomated task” suggests that
mind wandering requires working memory and executive resources.
However, although
evidence clearly exists to support the executive function viewpoint, I have
several concerns with the way in which the evidence was collected and the way
mind wandering is defined. For instance, much of the results discussed here
rely on self-reports and questionnaires. This runs into the exact same problem
that we discussed in last week’s blog post, namely a lack of consistency.
Without a rigorous definition of mind-wandering, the authors of these studies
risk invalidating their results. Specifically, subjects can choose to not
report TUITs or may not catch slight instances of TUITs. Furthermore, the
conception of what constitutes “mind wandering” may be different among
subjects, which serves as a further possible confounder of the results.
Furthermore, I found the meta-analysis of the evidence a bit troubling.
Although I know that this happens very often in research and can be very
enlightening, I think it is important to have normalization procedures in place
when one does such an analysis. For instance, Smallwood cites evidence from
several different times of studies, from SART studies to text comprehension.
However, he did not dwell upon the possibility that evidence from across such
different studies may require further processing before they could be compared
to each other.
In reply to
Smallwood’s 2006 paper, McVay and Kane (2010) provides evidence for a control
failure model of mind wandering. In accordance with this theory, mind wandering
should be prevented when “control is proactively initiated and maintained” in
response to task demands. Importantly, they distinguish between two types of
executive control, namely proactive and reactive, that are dissociable by
behavioral measures and association with different brain areas. The majority of
the evidence for McVay and Kane stems from studies done utilizing individuals
with differences in their propensity to mind wander. Specifically, the authors
provide what they believe to be counter-evidence for Smallwood and Schooler
(2006)’s executive function theory. They argue that if mind wander indeed
demands executive resources, then individuals with more resources (namely high
working-memory-capacity individuals) should exhibit more mind wandering than
low working-memory-capacity individuals. However, they found the opposite
effect: subjects with higher WMC clearly demonstrated greater executive control
and less frequent mind wandering across tasks. The authors argue that this
evidence serves to counteract the executive function resource theory presented
by Smallwood and Schooler (2006). Furthermore, the authors cite the default
network to provide evidence for their executive failure theory. The default
network is essentially a collection of different regions in the brain are
implicated in mind wandering (found to be especially active during mind
wandering). The authors reference the neuroimaging results from Weissman et. al. (2006), which found trade-offs
between control areas of the brain and the default network. Specifically,
reductions in PFC activation, which was interpreted to be reductions in
attention-control area activity, “reliably predicted lapses of attention”
(McVay and Kane, 2010). The authors cite this “anti-correlation” between
default network activation and executive network activation as evidence for
their view that executive control failures are responsible for mind wandering.
There are several
clear problems with the strategy presented above. Specifically, the association
of the prefrontal cortex with exclusively proactive executive control and the
anterior cingulate cortex with exclusively reactive executive control provide
an oversimplification of the functions of the PFC and ACC. Although the
association may definitely be present, there is no definite engram, and no
particular region of the brain is exclusively responsible for one task. Furthermore,
there are clear alternative explanations for PFC activation that were not
considered in the studies mentioned. Thus, one cannot make any decisive
conclusion without proper controls, which were also not described in the
papers. Furthermore, I found the association of working memory capacity (WMC)
with executive resource capacity very troubling. McVay and Kane (2010)
essentially utilize WMC as a proxy for the amount of executive resources an
individual has at his/her disposal. Although working memory capacity definitely
counts as an executive resource, it is not the only such resource, and a higher
working memory capacity does not necessarily imply the existence of greater
executive resources. Lastly, I found the authors’ discussion of the default
network as evidence particularly troubling. Again, the utilization of PFC
activation as a proxy for executive control without the proper controls
impinges severely upon the validity of their data. Furthermore, the conclusion
they drew from the results was also ungrounded. Indeed, the trade off between
“executive” areas of the brain and the default network could provide evidence
for the very hypothesis McVay and Kane (2010) are trying to disprove, namely
the executive function hypothesis. One can interpret this observed “trade off”
as an indication that mind wandering utilizes executive resources, thus leading
to lower activation rates in the executive areas of the brain (corresponding to
a higher level of activation in the default network). Thus, the evidence
presented has clear flaws.
The Choice
After evaluating Smallwood’s original
resource theory centered executive function viewpoint, McVay and Kane (2010)’s
evidence, and Smallwood’s reply, I must say that I am leaning towards the
executive function viewpoint. However, this is only in light of the evidence
presented for each theory and the qualifications the respective authors
mentioned. On a psychological level, however, I don’t think either theory really
tells the whole story. Rather, I think a hybrid of these two theories would
make for a complete description of mind wandering. Executive function and
executive failure models need not be completely mutually exclusive, as
Smallwood implied in his 2010 paper. Just as different types of mind wandering
exist (recall Giambra defined at least two: directed mind wandering and
involuntary mind wandering), I believe different mechanisms behind mind
wandering also exist. For instance, executive function may account for mind
wandering during certain conditions, such as those in which mind wandering is
beneficial for adaptation or creativity purposes, whereas mind wandering could
be due to executive failure in other situations. I would be very interested to
see further investigation into this topic matter and especially the
introduction of a hybrid theory.
Links to Neuroscience
Okay, after my rant on McVay and Kane
(2010)’s assumptions in their neuroimaging evidence, it’s clear that I think
it’s very difficult to biologically evaluate these two theories. Don’t get me
wrong, though; I think neuroimaging provides extremely valuable data. The key
is in the interpretation of the neuroimaging results; correlation studies are both
insightful and conducive to future studies. However, one just needs to be
careful to not form causal relationships on the basis of correlational studies.
In addition to neuroimaging studies, however, I think it would be interesting
to have a mouse model of these two viewpoints. For instance, the researcher
could create two separate populations of mouse, with the experimental group
being deficient in executive resources. One way to accomplish this could be
silencing a key pathway that is known to be important in information
processing. Furthermore, one could also create a third group of mice with
working memory deficits. We can then train these mice to perform a specific
task, and task efficiency could be evaluated across populations to see if a
difference exists between mice with access to different degrees of executive
resources. Furthermore, we could also compare the WMC deficient mice with the
executive resource deficient mice to see if the former’s task efficient
accurately predicts that of the latter. However, several hurdles must be first
overcome before this could be realized. For one, a pathway that is critical to
information processing needs to be defined and results need to be accurately
controlled for. Given the appropriate controls and experimental design, I think
this study could be very enlightening!
References
Image References
Got Questions? http://www.victesolin.ca/imported-20101026012429/2011/1/12/got-questions.html
Accessed 02/27/2013
National
Geographic. Daily News. http://news.nationalgeographic.com/news/2010/07/100701-science-animals-mating-mice-mouse-aphrodisiac-tears/
Accessed 02/27/2013
Illustration
Source. http://www.illustrationsource.com/stock/image/247949/an-illustration-of-a-woman-arriving-at-a-crossroads/?&results_per_page=1&detail=TRUE&page=9 Accessed 02/27/2013.
No comments:
Post a Comment