for discussion 3
Monicarodr
interested in it. However, from the selection I gave you in our
syllabus (which are uploaded here already), we have enough methodology
to discuss, so we will stick to those for discussion 3.
Have a happy weekend, week, and socialneuro reading!
Monica
UBT3
One thing that jumped out at me from these readings was that every
paper had a similar approach to testing the same theory…and then I
belatedly realized why—there is crossover between authors on every one
of the papers. Essentially what we are reading is the evolution of a
new idea into a more established idea over a series of papers between
2002 to 2006. It makes a lots more sense if you have a sense of this
and read chronologically (having said that, not what I did,
whoops). ;) So, as far as methodology goes overall I would say that
the theory is ultimately addressed from multiple converging
directions, making the evidence seem stronger. However, in the
earlier work, it is more exploratory in nature and thus the methods
and experimental design is weaker, as would be expected. I will focus
my comments on the 2002/2005 articles below.
From what I can gather, Turk and colleagues (2002) and Kelley and
colleagues (2002) were published approximately the same time, and I
take it that the Turk et al probably shortly preceded Kelly et al
since the Turk version was more of a short summary than a completely
designed study, and was set up using exploratory procedures and
general predictions. Later articles evolved to predict and test much
more specific hypotheses (to include distinctions between the dorsal
and ventral mPFC) whereas in these initial conceptualizations, just
distinguishing the existence and importance of a self-reflective
center/location was a major conceptual hurdle/accomplishment. Turk et
al (2002) only measures one subject with a cut corpus collosum to
establish differential activation of hemispheres for self/other to be
used as an initial evaluation of the concept. The authors evolve to
use the more sophisticated procedure of averaging measurement of
reaction changes over a group of 11 subjects, which constitutes a more
robust finding, as well as a different angle of examination. Mitchell
and colleagues (2005) are eventually able to round out the theory by
summarizing subsequent relevant findings in the field, explaining why
the effects observed might be seen (similarity to self activates the
self-region because it is simulating personal experience to explain
that of like-others). This study also uses a larger sample size (18
subjects) and included ratings of similarity to the self, which were
found correlate with the self-activation when emotional predictions/
theory of mind was required, but not for objective judgments of the
same faces. In later studies this concept was expanded into the self/
close other/familiar other paradigm and used to predict activations.
Stuart Daman
the brain. Two of the articles confused me a bit because they were so
disgustingly similar. Kelley et al. (2002) showed that processing of
information that is self-relevant resulted in activation in a unique
area, the mPFC; whereas self-and other-relevant judgments resulted in
activation in the lIFC (i made that one up). Then, Heatherton et al.
(2006) showed that this can be ever so slightly modified and extended,
showing that activation in the mPFC occurs in thinking of both the
self and intimately-close others, whereas the previous research showed
that not intimately close others did not result in activation of this
area. It was almost confusing because these articles we so similar.
This was initially evident in the fact that they share three of the
same authors, so it was obviously the same 'research team'. In
addition, the Methods paragraph subheaded "Imaging" was 100% EXACTLY
the same in each of these articles. That's awfully lame, if you ask
me. I understand that the two studies are remarkably similar, but they
are also each illustrating to importantly different points. Recycling
your own words may not qualify as plagiarism, but it does look shady
and sloppy, if you ask me. And that's about all I have to say about
that.
The other article that really caught my attention was the one by
Mitchell, Banaji, & Macrae (who also co-authored the aforementioned
two papers) (2005). By the time I get to the end of an article, I
often dread reading the discussion and it's sometimes senseless
rambling, but I was able to trudge through it in this article.
However, I feel this was largely in anticipation of reading the
Methods section (which I am and have previously noticed is often
placed at the end of paper in neuroscientific journals). In
particular, I checked the studies for indications of whether or not
they engaged in what Vul et al (2009) called the 'nonindependence'
error. As was the cause for their review, I could not tell for sure
whether they used the same information to choose voxels as to test
them, but it seemed to me as though they did. This alone brings into
question their results.
This aside, their discussion brought up several pieces of information
I found interesting. The entire incorporation or comparison of
simulation theory and "theory-theory" (a truly humorous nickname if
you ask me) approaches to information processing was neat, for one
thing. It's always nice to be able to prop two theories against one
another in research as well. It was interesting to me that the
evidence seems to support simulation theory when virtually all of the
research I've read till now about self and others suggested support
for "theory-theory" approaches. This makes me want to learn more about
simulation theory and search the behaviorally and survey based
literature for stuff about it and self and others.
I've also always found the idea of mirror neurons fascinating,
although I guess they're still kinda catching on: "so-called mirror
neurons" (p. 1311) as they referred to them. Where the main idea is
that mirror neurons are activated both when we perform an action as
well as when we someone else perform it, some research suggests that
this extends into simulating (i.e. mirroring) emotional reactions.
Then they say that this has been theorized as some to be a
"rudimentary system for simulating the minds of other people" (p.
1311). Is it just me or is this brutally obvious? Of course that's
what it suggests! And I find the notion of pinning down a specific
mirror system pretty cool, because this would indicate the mirror
operations may be more localized than the behaviors/feelings that are
being mirrored. It also raises the question of whether this system is
localized for different basic type of mirroring, such as behavioral
actions being mirrored specifically in the motor cortex and emotional/
affective things being mirrored in a specific part of the mPFC
(although I got lost in the talk regarding specific parts of the
mPFC). The authors then go on to comment on how less intimate forms of
interaction (e.g. phone conversing) reveal fewer cues to help in
inferring anothers mental state, and I couldn't help but take it one
step further than phone conversations in thinking about how difficult
it is to discern mental states via simple text. As a very basic
example, when you are reading this, how do you think I am approaching
it? Do you think I am senselessly rambling like someone who likes to
hear their own voice? (regardless of the fact that I'm not speaking
them) Or do I sound (look) like I'm angry and rambling about every
little thing in this research that pisses me off? Those are just two
examples, my point is that it is even more difficult to tell my
'mental state' from reading my text, which is even further removed
from face-to-face interaction than a phone conversation.
One last comment on this paper, what the hell is an "inverse
correlation"? I've NEVER seen anyone refer to one of these before in
my life. There are positive and negative correlations, positive and
negative relationships, and relationships and inverse relationships.
However, from my knowledge and background, "correlation" is NOT
synonymous with "relationship", too much more specificity is implied
in a correlation.
On a final note/question, why is it important to know/control for the
handedness of participants in fMRI studies?
Jenny Perella
in this week’s readings about perception of the self and others. It is
located on the medial surface of the cortex on each prefrontal lobe.
According to Beer, Shimamura, and Knight (2004, as cited by Pinel &
Edwards, 2008 in our coloring book), the medial prefrontal cortex is
involved in the regulation of social and emotional behavior,
specifically the monitoring our own behaviors/behavioral outcomes, the
encoding of self-relevant information, and the monitoring of others’
mental states, all of which is presented in our readings for the week.
The coloring book also gives a good review of the proposed functions
of the posterior anterior parts of the medial prefrontal cortex: the
posterior medial prefrontal cortex is the part that is thought to
“continuously monitor and control behavior” in order to “ensure that
the behavior is consistent with intentions”. The anterior medial
prefrontal cortex is thought to do this too, but is thought use more
complicated analyses, considering feelings and intentions of others
(Amodio & Frith, 2006, as cited by Pinel & Edwards, 2008, p. 216).
Though some of this week’s studies discuss other brain regions
involved in self-other processing, I will focus on the mPFC for
brevity.
Turk et al. (2002) kind of introduce us to the topic of self-other
processing in the brain. They discuss that face recognition relies on
structures in the right cerebral hemisphere, citing that damage to
these areas impairs people’s ability to recognize others’ faces. In
split-brain patients, the left hemisphere shows a recognition bias for
the self and the right hemisphere shows a recognition bias for
familiar others. Some studies suggest the right hemisphere could be
used for self recognition too, but brain imaging studies show that
highly self-relevant material activates left hemisphere, not the
right. In this particular article, Turk et al use the example of JW, a
split-brain patient, to illustrate. Accordingly, JW’s right hemisphere
showed a bias toward recognizing a familiar other, and his left
hemisphere showed a bias toward recognizing the self. The authors
conclude that both hemispheres are capable of self-recognition, but
the left hemisphere is important for execution of the process.
The study conducted by Kelley et al. (2002) shows movement in the
field from relying on hemispheric assessments to exploring more
specific regions of the brain implicated in the processing of self-
other distinctions. In this study, participants were imaged while
making judgments about trait adjectives under conditions of self-
relevance, other-relevance, or case judgment. Relevance judgments
compared to case judgments were correlated with activation of the left
inferior frontal cortex and the anterior cingulate. Self-judgments,
when compared to other judgments, revealed greater activation in the
medial prefrontal cortex, and both self and case judgments revealed
greater activation in the posterior cingulate. Selective activation of
the mPFC for self and other judgments (not case judgments) suggests
that personal-referential processing is functionally different from
other forms of semantic processing within the brain.
It is important to note, however, that both of these regions exhibited
decreased activity when all encoding trials were contrasted with
baseline; the activation discussed during self-judgments is referring
to less of a decrease from baseline than other/case trials. Raichle et
al (2001, as cited by Kelley et al, 2002) suggested the existence of
“an organized, baseline default mode of brain function that is
suspended during specific goal-directed behaviors.” If this is true,
it makes sense that baseline activity of the MPFC is highest at rest
and decreased when we are processing information about the self. I
would have liked to have seen more discussion on this point; the
authors seemed to just assume nothing strange or interesting about the
activation of a process (self/other judgments) being related to
decreased activity in the brain… it seems there is more going on here
that was not explored.
I thought the Mitchell (2005) article was the most interesting because
it was the only one to look at differential activation of multiple
regions within the mPFC, specifically the dorsal and ventral mPFC.
Mitchell proposes activation of the mPFC follows simulation theory,
which suggests that people infer the mental states of others by
imagining their own thoughts, feelings, and/or behaviors in similar
situations. If self-reflection is used to infer the mental states of
self-similar others, it is not surprising that a part of the ventral
mPFC was found to be activated during processing of self/other
similarity, given that this region has previously been implicated in
self-referencing tasks. To test the mPFC activation according to
simulation theory, the authors decided that the mPFC should be
selectively active when inferring others’ mental states, but NOT
active when inferring others’ appearance or physical location. The
mPFC should also only be active when inferring the mental states of
similar others, as simulation theory suggests that one does not use
one’s own mental states to infer the thoughts/feelings of others who
are dissimilar from the self. They test these theories by gathering
fMRI data for participants judging the mental or physical states of
similar and dissimilar others.
Mitchell et al found that their hypothesized brain regions, and
specifically the dorsal mPFC, were active only when assessing others’
mental (not phsycial) states, and that there was a correlation between
ventral mPFC activation and perceived similarity to the self. The
ventral mPFC was only marginally more engaged by mental states than
physical states. Thus, dorsal mPFC responds more to content (mental
state appraisal) and ventral mPFC relates more to familiarity (similar
vs. dissimilar to self).
Mitchell addresses the issue of decreased mPFC activation during self-
judgments from a different perspective than Kelley et al. As mentioned
above, past studies have shown only a relative activation of the
ventral mPFC (deactivation from baseline, but activation in comparison
to reference/case judgments). This study is one of the first to show
actual ventral mPFC activation above baseline, suggesting that this
activation is contingent on perceived similarity of others to the
self. Still, I would have liked to see more discussion on this point.
Heatherton et al (2006) basically replicate the Kelley et al (2002)
study, using intimate instead of familiar but impersonal others. They
show that the neural correlates of the self and others are distinct,
regardless of whether the other is intimate or not. Participants
viewed trait adjectives and were asked to judge if it a) described the
self, b) described an intimate other, or c) whether the word was
presented in capital letters. As in other research, making judgments
about the self relative to the personal other selectively activated
the mPFC. Though definitely contributing to our knowledge in this
area, this research was a step back from the Mitchell et al article
because it did not look at activation in both the dorsal and ventral
mPFC. It would be interesting to see if activation in these regions
differed.
The Beer (2007) article was interesting, but I thought it lacked depth
and explanation of theory. This may be because it was targeted to
readers of Cognitive Neuroscience, not psychology. Beer suggests that
the higher resting metabolic rate of the mPFC represents a default
psychological state of “chronic self-evaluation that helps people
consider their strengths and weaknesses when planning future actions.”
Again, this is interesting, but where did this idea come from? I would
have liked to have seen more discussion. Since there is decreased
activation when making self-other distinctions, could this suggest
that another region is more active and is responsible for self-other
processing? Perhaps the mPFC must be less active in order for this
other smaller brain region to trump the typically overarching activity
of the mPFC. Beer mentions the research of Moran et al, describing
that they wonder if the mPFC activity in self-evaluation is more
cognitive or emotional, thus more logical or biased toward positive
self views. Without more discussion of the underlying theory, I’m not
sure I buy this possibility of extreme bias in our typical processing.
Beer, however, suggests that mPFC is related to cognitive (self-
description) processes, and the ventral anterior cingulate is related
to emotional processing of traits (positive vs. negative). She briefly
discusses the cognitive/emotional nature of the default self-
evaluative processes.
Overall, in terms of brain activation in the processing of self and
others, it seems that the left hemisphere could be more important than
the right (Turk et al, 2002; Kelley et al, 2002), and that specific
regions involved in this processing are the mPFC and the anterior
cingulate (Kelley et al, 2002; Mitchell, 2005; Heatherton, 2006). The
neural correlates of self-judgments are distinct both from evaluations
of similar familiar (but not personal) others and intimate others
(Kelley et al, 2002; Heatherton et al, 2006). Self-other processing is
in general associated with decreased activation of the mPFC from
baseline metabolic levels; this is true in the dorsal mPFC, thought to
be involved in the processing of mental states (not physical states).
The ventral mPF, however, did show increased activation from baseline
in response to evaluations of self-similar others, independent of
assessing mental or physical states (Mitchell, 2005). The typical
higher baseline activation of the mPFC could be indicative of a
psychological default to chronically self-evaluate (Beer, 2007).
Lindsay Morton
With a focus on the methods of the studies for this week, I have to point out my disagreement with Stuart on one point. Although Heatherton et al. (2006) and Kelley et al. (2002) do report the same methods, I do not believe this to be a form of plagiarism but rather evidence that the researchers were attempting to do an exact replication (albeit with different stimuli: familiar unknown other vs. familiar significant other) to build upon and extend the earlier research. I think this is important because it shows the exact procedures and equipment used so that other interested researchers might be able to perform similar studies.
In addition, the work by Heatherton et al. (2006) does a great job in not only replication but they also used an event-related design, which is subject to less error. More importantly, the work employed an unbiased, a-priori region of interest (ROI), based on the previous work by Kelley et al. (2002) and by Schmitz et al. (2004). The Kelley et al. (2002) study did not do this, and the work could be criticized on the basis of the non-independent error. At the same time, through the methodologically stronger evidence provided by Heatherton et al. (2006), one might still be able to utilize the results of these 2 studies, which suggest that the medial prefrontal cortex (mPFC) remains activated during self-referential trait judgments but not during non-self social judgments (i.e., familiar, unknown others and close others).
Sadly, the work by Mitchell, Banaji, and Macrae (2005) is also subject to the non-independent error. Though this does call into question the legitimacy of the strength of their results, the work helps to focus direction for future research. For example, while reading the Mitchell et al. (2005) article, I started to re-think a project that I completed during my first year in Dr. Muraven’s lab. In this work, we were trying to determine whether imagining oneself engaging in self-control would have similar depleting effects as actually engaging in self-control. Work by Ackerman, Goldstein, Shapiro, and Bargh (2009) on vicarious ego-depletion came out after the failure of our work to produce any significant results. At the time, I was curious why their work, which asked participants to mentally simulate the perspective of a person either engaging in self-control or not utilizing self-control, showed this mental simulation depletion effect. It seems to me that it may have something to do with the area of the brain that was being utilized for the two tasks. In our experiment, we asked the Simulation Group to picture themselves engaging in a self-control task (i.e., resisting delicious cookies while hungry) or a comparable neutral task (i.e., resisting healthy radishes after eating a big meal). Although the study had some other methodological issues, our null results may have been due, in part, to the fact that we asked participants to evaluate self-relevant information in both cases. It would be great if we had access to fMRI (or other neurological) technology in which we could test how mentally simulating first- and third-person accounts of self-control activity may differentially activate the mPFC. Such evidence, combined with information about brain activity during actual self-control task performance, could help to discern how depletion occurs (e.g., the higher metabolism of the mPFC during self-relevant activities in addition to the energy needed by other brain areas utilized during the self-control task).
Likewise, two of the other papers for this week highlighted areas for future research. In the Mitchell et al. (2005) article, the importance of perceived similarity in judgments of social others points to social neurological work that could help examine how in-group and out-group biases occur within the brain. One could also test how tactics to decrease out-group prejudice might actually be decreasing the response of the dorsal mPFC (note: idea based on pg. 1309 findings). Another idea stimulated by the Beer (2007) article is a possible way to look at the divergent findings on the effects of mood on motivation. Specifically, the work of Moran et al. (2006) cited by Beer found that the mPFC and posterior cingulate are active during the cognitive aspects of self-evaluation but that the ventral anterior cingulate is active during emotional aspects of self-evaluation. In line with this, goal progress and motivated behavior are thought to rely on a cognitive evaluation of whether or not specific factors (related to the self and the situation) are appropriate for the expenditure of effort. At the same time, some work has found that mood and emotion can impact how individuals choose to expend effort in the pursuit of specific goals. If these two evaluations are separate (as indicated by Beer), I think it becomes important for the field of social neuropsychology to determine how these areas communicate and translate into real behavior. This could help to resolve the long-standing debate of the impact of mood on motivation.
Finally, the case study by Turk et al. (2002) provides preliminary evidence for the dissociation between self and other facial recognition. Though this is a single subject design of a split-brain patient and is therefore not greatly generalizable, the method of presenting morphed faces is an interesting one. I would be curious to see whether or not there is evidence for left brain dominance for the processing of self-relevant information in “normal” subjects. On a somewhat related note, I have noticed that all of these studies utilize participants who exhibit strong right-handedness. It is my understanding that handedness influences brain development and activation (i.e., brain lateralization). This raises a rather disconcerting question for me: do all of the results that we have found only generalize to right-handed individuals? Though this may seem like a strong position, I think it’s one that we should consider, especially seeing as about 10% of all people (about 600,000,000 individuals) are lefties.
Camille Barnes
self and others
This week’s articles focused on the idea that the self is processed in
a different manner than other people are processed in the brain. Self
processing seems to activate the left inferior frontal cortex,
anterior cingulate, medial prefrontal cortex (Kelly, Macrae, Wyland,
et al, 2002), ventral medial prefrontal cortex (Mitchell, Banaji,
Macrae, 2005), a region of the medial prefrontal cortex (Heatherton,
Wyalnad, Macrae, et al, 2006), and more generally the left hemisphere
in Turk, Heatherton, Kelley, et al, 2002.
I found it interesting how in the Mitchell, Banaji, and Macrae
article, that the ventral mPFC functioned almost directly opposite to
the dorsal mPFC. The ventral mPFC was active when using self-
reference to infer about similar other’s mental states and not active
when inferring about dissimilar others. The dorsal mPFC was active
when imagining dissimilar others, but not while imagining about
similar others. Usually, when you read articles they will state that
a certain region is active during a task. So, to have to parts of the
same region acting in an opposite manner seems surprising.
I did have a question about the Kelley, Macrae, Wyland, et al. study.
The concept behind this study was based on cognitive research that
found superior memory for self-relevant words, other relevant words,
and case distinction. But the most recent installation to this
research found that superior memory occurs for words related to
survival (above and beyond self-relevant information). It would be
interesting to gain information on brain activation to this survival
information in comparison to the other conditions to reveal more about
the memory effects. For instyance, does this survival information
activate the same brain region as self information, just to a greater
extent, or does it activate an entire different region of that brain,
that leads to superior memory.
The Mike or Me? Article proposed that there was a left recognition
bias for self distinctions, and a right recognition bias for familiar
other distinctions. However, I was not completely convinced of the
argument, because the familiar other they used was the doctor/
experimenter. Wouldn’t it be possible that rather than there being a
left bias for self-recognition, that it may be a left bias for highly
familiar information? I would be more convinced of the argument if
the “familiar” other used in the study was a best friend or spouse.
Either way the article does seem to show that less familiar people are
processed in a different way than the self, I am just not sure if this
is due to self vs. other distinction, or a level of familiarity
distinction.
David Dinwiddie
For the Kelly article the methods were presented after the results which I have never seen before. In this study 24 participants were used. 2 of these participants were eliminated from analysis due to “technical difficulties with fMRI data reconstruction.” A third participant was eliminated due to excessive movement during imaging. I am glad they eliminate participants for this but after these eliminations they were left with only 21 participants. The apparatus was identical to that in the Heatherton article. (likely the exact same apparatus.) The methods in this study were also identical to that in the Heatherton article except that they were now making judgments about President George Bush instead of a friend.
In the Kelly article I had a problem with their behavior results. First they should have presented their mean latency time for self judgments. It seems that the times may not have differed that greatly in latency between those yet they were significant due to such a large number of trials. Overall the behavioral results were of no surprise to me in these two studies. One should recall adjectives better if they make a self judgment on them as opposed to an other or case judgment. It is interesting that the MFPC has greater levels of activation when making a self judgment that when making an other or case judgment showing that not all judgments are occurring in the same way.
Jennifer Vosilla
Oy! Sorry I'm late!!
Overall, the articles seemed unanimous in agreeing that the medial prefrontal cortex (MPFC) is active while processing self-relevant information as well as having an unusually high baseline metabolic rate in comparison to “many other brain regions”. Beer (2007) mentioned this high baseline comparison, but I didn’t like how vague she was in that she didn’t specify exactly how many or which brain regions also had high baselines. It is hypothesized that this baseline suggests that in general there is a constant, automatic evaluation of self. Beer (2007) explains the findings of another study and its implications, but then critiques it suggesting that a neutral trait is needed to explore the positivity bias. Personally, I think that critiquing another’s work and giving suggestions for future research does not really contribute that much. I would have liked to see her actually implement her suggestions.
Turk, Heatherton, Kelley, Funnell, Gazzaniga and Macrae (2002) used a split brain patient (JW) in their study. Using 11 facial photographs varying in degrees of self-likeness and familiar other-likeness, JW was asked to indicate whether the photograph was himself (‘Self’ Recognition Condition) or to indicate if it was Dr. Michael Gazzaniga (‘Familiar Other’ Recognition Condition). The authors found a double dissociation supporting their hypothesis that the right hemisphere has a recognition bias for others and the left hemisphere has a recognition bias for self. I felt it made their argument stronger because they then replicated the double dissociation 3 more times using different ‘familiar other’ photographs (Clinton, Bush and personally known individual).
Kelley et al. (2002): Essentially, this study was searching for whether processing of self-referential information uses a specialized area of the brain, which is speculated to be in the mPFC. I felt the article had one of the strongest theoretical backgrounds as it built upon and replicated evidence from cognitive psychology that there is increased memory for self-referential information. The authors also cited and described many other studies that supported their finding for support that the self does have a specialized area and that even other semantic-based processing does not yield the increased activity in the mPFC. Heatherton et al. (2006) builds upon this study by using the same procedure only substituting the other with a friend judgment. It has been theorized that close others are perceived as actually an extension of the self and this is what the authors were looking to test and found that this is not true in reference to significantly activating the mPFC for close-others.
Mitchell et al (2005) may have been my favorite article because I enjoyed the idea of being able to pit two opposing theories against each other and using the fMRI to find support for only one theory. The authors found support for the simulation theory that people use themselves as a reference point (activity in mPFC) when attempting to infer the mental state of others, but this is dependent on whether the other is perceived as similar to oneself and only when mentalistic aspects are involved. However, the article points out that there are limitations to having “either-or theorizing” as humans are just plain complex.