Skip to main content

Burden of proof: does neuroscience have the upper hand?

As an undergraduate, I took several introductory level philosophy classes while
majoring in neuroscience. Some of it I could appreciate and most of it went
over my head, but a thought that kept nagging me was, “haven’t neuroscientists
solved all of these issues by now?” It was only after I had worked in
neuroscience laboratories for a few years that I began to realize just how
qualified all of our statements had to be due to the plethora of limitations that
go along with any result. I began to wince anytime I heard someone use the word
“proof” (only salesmen use the term “clinically proven”, but don’t get me
started on that…). It seems clear to me now that, for the most part, natural
scientists, social scientists and humanities scholars are really all working
toward the same goal just in different, albeit complimentary ways. At the first
“Neuroscience, ethics and the news” journal club of the semester, Lindsey
Grubbs, a PhD student in Emory University’s English Department, facilitated our
discussion about a topic that she has previously written about for this site.
The main focus was on what role neuroscience can and should play in answering
questions that have long been in the realm of the humanities and how these
results should be communicated to the general public.

From the Daily Mail Online

At the center of
our discussion were two papers about the effects of reading that each created
quite a stir
in the popular press. First, Gregory Berns
laboratory at Emory reported on a study that they had conducted which was aimed
at determining how a good book can leave such a lasting impression (1).  They reasoned, “It seems plausible that if
something as simple as a book can leave the impression that one’s life has been
changed, then perhaps it is powerful enough to cause changes in brain function
and structure.” Berns and colleagues asked undergraduates to read a novel over
the course of nine days and had them come into the lab for a short
resting-state functional Magnetic Resonance Imaging (fMRI) scan each morning
during that time. They utilized an internal control model where each
participant was also scanned prior to reading and after finishing the novel.

Dr. Berns and his
colleagues found significant short-term changes in activity levels in areas
that had previously been associated with “story comprehension” and “perspective
taking” – the “left angular/supramarginal gyri and right posterior temporal
gyri” – and somewhat persistent changes in somatosensory cortical connectivity.
They interpreted the latter results as a possible substrate for the phenomenon
of “embodied semantics,” where the brain’s somatosensory processing machinery
can be recruited by just the thought of performing an action or experiencing a
sensation (2).
Importantly, nowhere in the paper do the authors put a valuation on these
changes as “good” or “bad” yet headlines such as “Brain
function improves for DAYS after reading a novel
” appeared. Increased
connectivity between discrete brain regions does not necessarily mean that
brain function is improved. Is this simply a result of a positive bias in favor
of reading? If the same neurological changes were found in chronic drug abusers,
I doubt they would be interpreted as improvements.

From ScienceDaily

The second study (3),
published in Science by Kidd and
Castano (2013), took a very different approach and also received a great deal
of attention (positive,
and borderline
). Here, the researchers from The New School
set out to determine if literary fiction – defined as award-winning and/or
canonical rather than best-selling – alters measures of Theory of Mind, the ability to
infer and understand the mental states of others. To do this, they put subjects
through a battery of tests aimed at measuring affective and cognitive Theory of
Mind after reading either literary or popular fiction or nothing at all. Kidd
and Castano conclude their article by arguing that their results highlight the
importance of reading literature in contrast to the controversial US Common Core standards which de-emphasize
reading fiction
in secondary education. Slate’s Mark Liberman has written at length
about the shortcomings
of this study’s design and its perhaps over-reaching interpretations. His piece
was titled “That study on literary fiction and empathy proves exactly nothing”
– I couldn’t agree more (but I could say that about any paper). Liberman’s
somewhat aggressive title may have been in response to Zach Schonfeld’s “Now
we have proof that reading literary fiction makes you a better person” (shudder).

Hyperbole aside,
there are two issues here that deserve attention. First, do we really need neuroscientific evidence in order
to promote reading, and in particular, reading great, canonical works? Even
hardcore, card-carrying reductionists would likely agree that a lack of
biological evidence for the benefit of reading does not necessarily mean it
isn’t good for you. The risk is that the public comes away from these articles
thinking that now that neuroscientists have weighed in, the debate is over.
Also of note, Kidd and Castano never use the word “brain” (or “neuron”,
“neural”, etc.) because this is a psychology study, aimed at understanding the
effects of reading on how the mind
works. Neuroscientists could certainly look for neural correlates for the
psychological changes exerted specifically by literary fiction on the brain but
even if they were not able to find anything that would not mean that the
changes aren’t real. An underlying issue here, outlined by Lindsey and
discussed by the group, is that natural science results – and in particular
those from neuroscience – are highly persuasive to the public (4, 5).
Here, the media largely reported that these studies proved the benefits of
reading, which hardly seems like a controversial topic. However, Common Core
standards are quite controversial
and the amount and type of reading (i.e. fiction vs. non-fiction) are hotly
topics. While neuroscientists and psychologists certainly could
weigh in here, there is a concern that their results may be more influential
than perhaps they should be.


Second, a problem
that arose with several of these popular press articles is the attachment of a
value judgment to the changes that scientists reported. As mentioned above, it
is likely that the reason that words like “improvement” were added to describe
these neural changes is because reading is already seen as a positive
influence. Internet pornography addicts probably would have had a similar change
– if it existed – labeled as a “pathological rewiring.” Perhaps this is a
remnant of the brain being thought of as a muscle that needs to be exercised.
In that analogy, any increase does seem like an improvement but that obviously
is not always the case (for example, uncontrolled excitatory activity can lead
to seizures).

In reality, most
of the things we do and see and feel can probably change our brains in some
way, for better or worse – but whether or not neuroscientists alone are able to
find these changes doesn’t actually prove


1.         Berns GS, Blaine K, Prietula MJ, &
Pye BE (2013) Short- and long-term effects of a novel on connectivity in the
brain. Brain connectivity
2.         Aziz-Zadeh
L & Damasio A (2008) Embodied semantics for actions: findings from
functional brain imaging. Journal of
physiology, Paris
3.         Kidd
DC & Castano E (2013) Reading literary fiction improves theory of mind. Science 342(6156):377-380.
4.         Caulfield
TR, C. Zarzeczny, A (2010) “Neurohype” and the Name Game: Who’s to Blame? AJOB Neuroscience 1(2):13-15.
5.         Weisberg
DS, Keil FC, Goodstein J, Rawson E, & Gray JR (2008) The seductive allure
of neuroscience explanations. Journal of
cognitive neuroscience

Want to cite this post?

Purcell, R. (2014). Burden of proof: does neuroscience have the upper hand? The Neuroethics Blog. Retrieved on , from


Emory Neuroethics on Facebook