Personality Psychology (740)

         a blog


Marshall & Brown (2006)

Filed under: Uncategorized — drcb @ 5:31 pm
43 Comments to “Marshall & Brown (2006)”
  1. Sergiu Barcaci says:

    This study examines how the aggressiveness trait correlates with social provocation using the TASS model. As I started reading I instantly started thinking about potential issues that could arise during this study but I was satisfied by the end in seeing that these issues were accounted for or at least addressed. Marshall and Brown utilizing both moderate and strong provocation for both the low TA and high TA participants does build a strong study but what about the people that tested in the middle ranges of the BPA scale? While they do address this, their response is that society does in fact split itself with that dichotomy- you’re either labeled as aggressive or not. In a way, that is a good enough explanation, especially for a study that is very much in the preliminary stages in its field, but future studies should account for this, and also include a significant number of both genders even if it apparently doesn’t have that strong of an impact.

    Regarding the participants, the authors don’t specify whether they were psychology students so they potentially weren’t, but if they were this goes back to the discussions in class about how psychologically aware are the participants going into the study. The authors taking this into account and applying it to study 2 and 3 led more credibility to the study in my eyes. In study 2 they test to see just how aware a “layperson” is about how TA functions within a threshold (which resulted as expected in accordance to the TASS model) by gauging how they think a person will react in a hypothetical story. Study 3 is practically the same as study 2 but the participant is told to think of themselves as the person in the story and how will they react. These results did differ slighty (while still being the same general idea of high TA= quicker to anger) and I think it was because of the self-image issue. People think more highly and critically of themselves so they are more likely to say they won’t lash out even if they would as opposed to easily placing the same judgement on a made-up person in a story as they did in study 2.

    Going back to study 1, the results also could have been skewed due to not having to actually administer the loud noises to a physically present person- or at least did it like the Milgram experiment. The participants also being assessed for anger prior to aggression could have correlated to these answers as well, but the authors thought of acknowledging this. This study did account for many of its potential limitations and did prove that traits can be brought out by certain situations, while the severity of the situation also playing a key role in determining if the trait surfaces.

    • Dakota Egglefield says:

      Hi Sergiu, I was also a bit irked at the use of participants at each end of the spectrum (high aggression and low aggression). I agree with you that it was basically to prove a point, but in class we have repeatedly talked about how the majority of people fall somewhere in the middle on most traits. I understand that we have to study pathology (for lack of a better word here) in order to understand typical personalities since each trait falls on a spectrum, but I feel that they could have strengthened their study by also including people who fell in the middle of the spectrum, as most of us do. It would be interesting to see how high or low provocative situations affected those who were moderately aggressive and if there were any surprising findings.

  2. Scott Ewing says:

    I’m glad the authors clarified their use of a pooled error term in the footnotes of this article; I teach stats, and I was ready to go on a rant on the unethical use of multiple analyses to boost their p-values. But even if the stats are legit, the way the authors present them in the context of their hypotheses is extremely biased and misleading.

    For example, in Study 1 they had three analyses that were non-significant. When this contradicted their hypotheses, they still presented them as “partially confirmed”; “the means were patterned as predicted”, and variables “did tend to” fall where they expected, but the “effect[s] fell short of significance” (the p-values were .08 and .11). When a non-significant result supported their hypothesis, the numbers fell clearly “as predicted” (p>.10, which I’m just going to guess was .11). In both discussion sections, the authors speak as if all of their predictions were confirmed. “These conclusions were more clearly supported for overt aggression than for anger, but the pattern was generally the same.” I mean, c’mon.

    Let’s say for a minute, though, that their model is 100% correct and some of their results didn’t reach significance because of small or weird samples. Their next 2 studies demonstrate that a) laypeople are generally already aware of this pattern with trait-aggressiveness, and b) they’re even somewhat aware of how this model affects their own behavior (again, wacky presentation of stats, but the authors themselves conclude that people can predict this pattern for themselves). So, who is this benefitting? What impact does this have on the field? It looks like this series of studies was a considerable time commitment, so I fully understand the desire to publish their results. But their language implies a revolutionary new model that’s going to change everything; I’d have enjoyed reading it a lot more (it’s very well-written) if their presentation was more like, “hey, look at these neat results.”

    Clearly the article has me feeling aggressive. I rank pretty low on trait-level aggressiveness, so this must be a strong provocation for a geeky grad student.

    • Sunghee Kim says:

      Hello, Scott. First, thanks for your opinion. I do agree with you the ideas which the article itself used misleading terms which made the readers confused. And also there were three studies the researchers ran and the ratio of genders was very uneven. From my perspective, I was unable to find any definition of aggression/anger and researchers did not define these terms deeply. There were many situational/environmental factors besides just having aggression. Even if the results proved to be significant but can we really able to say that those results are significant? Even there were few things did not make sense in the article but I found it very interesting since I was able to retrospect my life why I tend to show my aggression even in little thing.

      • Gregory Rosen says:

        Hi Sunghee, The closest thing I found to a definition of aggression was “consider what it means to say that someone is an aggressive person. Commonly, this means the person reacts aggressively to a small affront. In contrast, a laidback, easygoing person requires a higher level of situational provocation before becoming angry. In more formal terms, we can say that trait aggressiveness (TA) reflects a heightened sensitivity to situational provocation, such that less in the way of provocation is needed to evoke aggression-relevant reactions in a person who is high in TA than in a person who is low in TA. As a consequence of these different sensitivities, TA will play its most important role in situations of moderate provocation” (Marshall & Brown, 2006, p. 1101). Do you think that description is insufficient? What do you think is a better definition?

      • drcb says:

        I think they were operating under the standard social psych definition of aggression which is, behavior intended to harm another who is motivated to avoid the harm.

  3. Sophie Schiff says:

    My main thought when reading this paper was how this research fits into the fundamental attribution error literature (a concept we touched on briefly in the last class). The fundamental attribution error is typically described as a person attributing a given behavior to an individual’s personality and failing to account for situational factors (Ross, 1977). This concept is interesting to consider in the context of the Marshall and Brown article because of the way the study was designed. Unlike in fundamental attribution research, where evaluators are asked to describe a given behavior retrospectively, study 1 of this article measured individuals’ reactions as a result of their own personality traits and situational provocation and in studies 2 and 3 participants were asked to predict an individual’s behavior, given information about his/her personality and the situational context.

    As I read this article, particularly for studies 2 and 3, I wondered about the authors’ conclusions about people’s inherent understanding and application of the TASS model and how this inherent model fits in with the research on fundamental attribution error. The authors acknowledge this concern in their implications section and claim that people “automatically adjust for situational strength (p. 1111)” but I do not find their explanation satisfying. If rather than predicting a behavior from a simulated prompt, participants were asked to evaluate an individual’s action or behavior without any knowledge of personality traits or situational factors, would participants demonstrate this same type of understanding or tendency towards an interactionist model? Fundamental attribution research would indicate that the answer to this question is, “no”.

    For example, if in study 2, rather than being asked to predict an outcome based on a described situation and information about the individual’s personality, participants were told that, “Chris was in a supermarket and was hostile, irritated, angry, etc.” and were asked to describe what they think was happening, would participants incorporate both situational and personality factors (automatically adjusting for situational strength)? That is, would participants respond something like, “Chris is a slightly irritable person who was likely cut in line at the supermarket that day” or would participants fall prey to the fundamental attribution error and claim that “Chris is an aggressive individual who likely was acting completely irrationally”? I wonder if the study design, particularly with the explicit description of the actor’s personality and the situational factors, led participants towards a certain prediction, as opposed to if they were forced to evaluate a given behavior in a more open-ended manner, which would likely lead to the fundamental attribution error and the incorrect emphasis on personality traits rather than the consideration of situational forces & the strength of those forces.

    • Sara Babad says:

      Hi Sophie,

      I actually had the same thought as you did, specifically in regards to the fundamental attribution error that we briefly discussed last class. I too, wonder if by giving situational context, the authors were priming the participants to take context in to account, leading to the implicit understanding of others’ actions that the authors describe. I think an interesting follow-up study might be to look at whether people’s judgments about others is, in fact, impacted by how many (if any) contextual cues are available. For example, if an acquaintance makes a rude comment when going through a stressful life situation, we are likely to be more understanding than if a stranger makes a rude comment. But, this difference could be due not to knowledge of context but rather to emotional attachment to a person making us more forgiving. A possible study might look at an unknown person with no context provided, an unknown person with context provided, and an acquaintance.

  4. Sara Babad says:

    Marshall & Brown offer a comprehensive understanding of how traits can be elicited based on the situation, specifically explaining that moderate intensity/impact situations are best suited for understanding how those high and low in a particular trait are likely to respond. After reading the article, I was left with the following thoughts/questions: 1) Perhaps it’s my background in the study of psychopathology, but I wonder if part of Marshall and Brown’s TASS model – certainly regarding anger – is dependent on emotional stability/regulation. The authors describe how medium impact situations will elicit an angry response from those high in trait anger as compared to someone low in trait anger. And they separately measured internals feelings of anger and overt aggressive behaviors, with stronger interactions found for overt aggressive behaviors than for internal feelings. To me, this indicates that someone might feel higher internal anger but not always respond aggressively, even if they are high in trait anger. And this might be described as someone who is more emotionally stable – who may get angry more easily but who is able to control their responses better. My question would then be – to what extent do you think emotional stability underlies many of the situationally-dependent traits? And what happens theoretically if emotional stability is itself situationally dependent, like any other trait. Is anger a special case because it likely falls under the umbrella of “neuroticism/emotional stability”, or is its seeming dependence on emotion regulation capacities applicable to a great many traits? Or perhaps I am confusing emotional stability with executive function capacities, like ability to inhibit and self-regulate, depending on the required situation. I wonder what your thoughts are.
    2) My second thought was about the excellent point the authors make regarding measurement error in studying personality, namely that how we study impactful situations can completely change the apparent relationship between traits and situation. This made me wonder if all fields are subject to this kind of error or if the study of personality is more susceptible to it because of the ambitious nature of what personality research is trying to do. Chapter 3: Traits and Situations, also from our assigned reading, states that in the study of personality, at some point, one can go in circles trying to parse out the relationship between traits and situations, getting lost in theoretical murkiness that leads to a dead end. This relates, too, to my first thought above. Is there something unique about personality psychology that makes it so theoretically “sticky”, so to speak?

    • Chen Tiferet-Dweck says:

      Hi Sara,
      Regarding your first thought, I think we shared the same concerns but you put it in better words. Basically you think that emotional stability type traits are more situational dependent than other traits? Am I right?
      And can we really separate emotional stability from executive function capacities? Isn’t emotional stability depends on our ability to inhibit and self regulate?

    • drcb says:

      Your ideas remind me of the concepts of anger-in(tendency to suppress anger), anger-out(tendency to express anger.

      You also bring up the issue of studying atypical events in life (such as aggression) versus studying everyday events. It seems social psych research is often geared toward studying unusual events, rather than the mundane.

  5. Sunghee Kim says:

    I found this article very interesting and was able to see myself better because it made me realize how I got into fights with people. I think of myself as a person who can be aggressive. There has been a time that little provocative situations triggered me to express my aggressiveness and anger. For example, I can describe myself as a cup filled with water. When there is 95% of space filled in a cup, water would not overflow; however, when you try to fill it up with more water into the cup, water will spill all over soon. It does not need much effort to do it but just a little bit more of water pouring into the cup. A person with low TA would be totally opposite of an almost filled-up cup. If I were a person like a cup with 95% of water filled, a person with low TA would need much more water to fill up his or her own cup before reaching to the point of being aggressive or angry. That is the main difference between low TA and high TA.
    In this study, it did not identify whether individuals with high TA were affected by genetic, situational, or environmental factors. For people living in a city like Manhattan, they may get upset easily due to factors like busy traffic, unkind drivers, noises, high expenses, and others. These environmental factors would be likely to impact on individuals with high TA. And yet, “Someone is aggressive” does not mean that the person is always angry. Rather, this person is able to easily express emotions, such as uneasiness, anger, and aggression, when a situation is given.
    Overall, there were fewer than 100 subjects from the University of Washington (UW) that participated in the study, excluding the study #1. My question would be using participants from only one university would be accurate to generalize to the population? (study #1) I would like to know whether this sample size would represent for the overall population. Hence, there were more female participants in the study, and it would influence the results. Not only that, the study #1 only had 1,101 women participated in the study. Would the result be the same when more men participated, or only be influenced by women?

    • Katherine Chang says:

      I like your analogy of the cup filled with water. Despite all my questions in my post, I still think I am in agreement of situational factors playing part and acting as a threshold. It is very logical and as in Sunghee’s analogy, it is also intuitive. But who knows, maybe I’m more prone to considering situational factors because my parents are from an Eastern society or because I’ve learned to consider the situation first instead of jumping to conclusions.

      Sunghee also brings up a good point of the representativeness of the sample. They were all women in the first study and while they said there were no gender effects in the other studies, it just makes me wonder. Also for the University of Washington, would there be a difference if the study took place in NYC? Or with different age ranges? I felt like I was much more reactive when I was younger. Ultimately, I’m curious about the internal situation on the situation, if you will. Would increased insight (via therapy potentially) help someone moderate their actions?

    • minjung park says:

      Hi, Sunghee. I like your idea. Especially, I agree with your opinion that the author did not investigate clearly the surrounding environment of the participants such as genetic or where they live. I also think if the author researched by limiting such details, I think there might be a different result. Besides, another problem about this article is that the authors used an unbalanced gender ratio and only UW student. I think these could also have affected the results in some ways. Once again, thanks for sharing your opinion.

      • Victoria Fairchild says:

        I had similar concerns about the participant pool, especially for study #1. A lot of generalizations are made in the discussion etc… regarding how people with high and low aggression react in varying “levels” of stressful situations. However men and women typically express emotion, and aggression in particular, differently both in word and action. I think it would be interesting to see what a more balanced sample’s results might look like.

    • Scott Ewing says:

      Love love love this analogy. I was imagining something similar but couldn’t come up with the right image, so I wondered if I could build upon your analogy to incorporate some of the other thoughts that have been mentioned.

      Maybe the cup is baseline Anger; when it overflows, the person becomes aggressive. Some people have more baseline Anger than others, but let’s say on average people’s cups start at around 50% full.

      The strength of the faucet that fills the cup could be influenced by Trait Aggressiveness, situational valence, and any number of other factors that have been mentioned in the posts above and below. If you’re low on TA, a moderate provocation might only “drip” into the cup, unlikely to overfill quickly. If you’re high on TA, the faucet gushes. And as Kathy, Sophie, and a few others mentioned, things like lack of sleep, high stress, sensitivity to criticism (vs other provocations) could open up the faucet more strongly for one person than it would for another, or on one day versus another. Dunno if that makes any sense.

    • drcb says:

      High TA individuals would be assumed to be partially so due to genetics, since aggression has shown to be heritable ~65%. Then the rest necessarily stems from environment. And here in this study, the effects of immediate environmental conditions is shown.

  6. Katherine Chang says:

    Marshall & Brown’s “traits as situational sensitivities” model (TASS) at first seemed very intuitive and logical to me. I like the idea of a threshold as that is the way I think of many things in life. However, as I read through the study, their definitions, statistics, and their conclusions, I had a couple concerns as well as some of my own questions. They posited that people with high TA would need less provocation in moderate as compared to those with low TA. Then when comparing moderate to strong provocation, those with low TA would have a greater increase of aggression as compare to those with high TA. I understood their results and the patterns and statistics aside, I got caught up with wondering more about the personal situational factors aside from the physical situation that was being produced. For example, how would someone with high insight and self-control react? What about if they’re running low on sleep or other potentially confounding factors? On an aside, I think aggressiveness (or how it manifests) can change with age and maturity which somewhat contradicts the idea of it being stable. My only support, however, to this claim is that when I took the Buss-Perry Aggression scale, it made me reflect that when I was younger, I had more difficulty controlling my anger (and maybe by proxy my aggression) than I do now.

    This brings me to another question about the author’s ideas about aggression versus anger. And also, in general, what is the correlation with the Big 5. Would someone high in neuroticism be more prone to aggressiveness? But then if they are high in agreeableness, would that negate that tendency? Where does anger come in?

    Regarding Studies 2 and 3, it reminded me of what was mentioned in class on how in Western societies attribute behavior to personality compared to Eastern societies that tend to explain behavior in terms of situational factors. How might this play out in this study? What if someone thought through the hypothetical situation of the person giving the feedback. Maybe this person had a bad day and is being more critical with his grading. I would think that hypothesizing about the situation (insight) would effectively reduce aggression/anger. But maybe that only happens with people that are prone to low TA?

    • Sophie Schiff says:

      I also thought about the personal situational factors that could be relevant in the outcome. Particularly, I was thinking about the relevance of the provocation to the individual. Perhaps there were differences in how people weighed the negative feedback on their essays in study 1. There are many other factors that could be involved in the response, like acceptance of criticism. Given two individuals who are low in TA, one could be highly sensitive to criticism and the other could be very comfortable with and open to criticism typically. How might these two individuals differ given the strength of the situational provocation? I understand that the authors were focused on differences based on existing trait level, but I think the relevance of the provocation to the individual would have been interesting to explore and consider in analysis as well, in addition to the other factors you mentioned (like personal level of insight, self-control, etc.).

    • Crystal says:

      Katherine, I think you bring up a great point about situational factors that could impact the findings, such as sleep deprivation. I’m sure factors such as sleep, diet, exercise, and stress level impact aggressiveness. I could see aggressiveness declining with age and maturity, but then again, there are many adults with high aggressiveness. In fact, I have seen many recent examples of adults with road rage in the news. Maybe it is stable for some or increases with stress associated with work, parenting, etc. Lastly, I agree that reframing negative thoughts (i.e., thinking to him/herself that the other person is having a bad day and is therefore more critical) would greatly reduce aggressiveness. It would be interesting to conduct a study in which level of aggressiveness is measured, then a cognitive behavioral intervention is implemented to teach how to reframe negative thinking, then aggressiveness is assessed again to see if it is reduced.

    • drcb says:

      Aggression tends to correlate negatively with agreeableness and conscientiousness, and positively with neuroticism (e.g., Ang et al., 2004).

      There are some lifespan developmental mean-level trends in aggression too. Toddler years are the most aggressive for most (but they don’t do much damage because they’re small). In the preschool and early elementary years, physical aggression decreases, and verbal aggression increases. 15 – 35 yrs old is the peak of violence. There’s also the issue of heterotypic continuity, whereby aggression manifests differently depending on age, but the tendency is stable.

  7. Chen Tiferet-Dweck says:

    Marshall and Brown (2006) present the TASS model which suggests that trait is best shown in situations that provide just enough provocation to trigger the reactions associated with that trait. Accordingly, they claim that the relationship between traits relevant reaction and situational provocation are not necessarily linearly connected and that the pattern behaves differently in different levels of situational provocation. They showed that people with high TA (trait aggressiveness) present aggressiveness for only moderate provocation. In addition, they showed the increase of aggressiveness from moderate to high provocation is smaller than the increase of aggressiveness from non-provocative situation to a moderate provocation. They concluded that when manipulating trait reaction we should use at least three levels of situational provocations to capture the real nature of that trait. I wonder if this dependency on situational provocation levels will be found in other traits as they claim or, perhaps some traits are more situational dependent than others? For example, I can see how the TASS model will fit traits like talkativeness, energy, courtesy, and I can see how we can manipulate the situational provocation levels to trigger these traits. But what about traits like trust and honesty? can we apply the TASS model to them?
    Another point that I wanted to discuss is the gap between the subjective self-report results (in studies 1, 2, & 3) vs. the objective behavioral results in study 1. The researches tested their model within each group separately. When examining self-reported assessments between groups subjects tend to predict or perceive that high TA people will be more aggressive than low TA people under strong situational provocation. However, the objective measurement of aggressiveness in study 1, suggests that both low and high TA subjects behave similarly aggressively under strong situational provocation (as predicted by the TASS model). These findings emphasize how important it is to have objective measurements as well. Subjective measurements are more vulnerable to subject biases and they can suggest different results than objective measurements.

    • Shira Russell-Giller says:

      Chen, I share you skepticism about self-report questionnaires, and definitely lean towards interpreting self-reports with a grain of salt. I appreciated you pointing our the discrepancy of findings between study 1 and study 3, namely, that in study 3, high TA participants expected to be more angry than low TA participants when faced with strong provocation. Since the actual experiment in study 1 indicated that both high and low TA participants has similar levels of aggression in reaction to strong provocation, perhaps the discrepancy is a self awareness issue. When faced with a hypothetical situation, perhaps the low TA patients were not able to accurately describe their reaction since a hypothetical situation isn’t enough to get them agitated – so they really don’t think they would be agitated in reality.

  8. Bengi Sullu says:

    Marshall and Brown (2006) present TASS (Traits as Sensitivities to Situational Strength) Model that posits traits to be functioning in a “threshold-like manner” (p.1100) in the context/presence of surrounding situations, which they conceptualize as “situational provocations”. Being high on a trait will mean that less situational pressure or provocation will be needed to generate response from that person compared to being low on the trait. The previous studies looked at the implications regarding aggression but this study’s strength is the original attempt to test that to bring out the individual differences relating to the level of one’s TA in relation to strength of the reaction shown to the provocation, moderate situational provocation is the most effective. Study 1 sets out to explore this, but I was surprised by proposed studies 2 and 3 before I went into reading them as their review did not seem to touch upon the subjects’ awareness of the link between aggression and situational provocation. In relation to the design of the first phase of the study 1, the fact that the process of evaluation and being evaluated takes place through the mediation of the experimenter made me think of the credibility and authenticity of the process of being evaluated from the perspective of the participant (which authors recognize in their limitations regarding the aggression part). I also think that assessing the mood through questionnaire that comes after the procedure might not be enough to capture the emotion at the moment they face the feedback. That’s why I was particularly curious about the result of the analysis about anger and interested to see that except for the comparison of moderate and strong provocation in low TA participants, comparisons of the other conditions yielded significant difference. In relation to Study 2, the predictions of the TASS models is confirmed according to the authors, but I think there is a problem associated with assuming what person thinks another person feels in a situation overlaps with what they themselves might feel. Study 3 tries to find out by asking participants to predict their own reactions in provocative situations but this embodies an assumption that equates not only perceived but also stated awareness with actual awareness, which in itself might point to different processes in high in TA and low in TA people that requires further investigation. In relation to the frameworks used, the authors underline the interdependency of traits and situations, I found myself thinking that from an environmental perspective, interdependency does not only imply that traits and situations interact but that in the making of the trait there will be ongoing situational contribution that makes the trait. In that sense, I think that subjects’ categorization of high and low aggression through an application of Buss-Perry aggression scale might needs to be supported by an additional scale or measurement, especially since participants have completed that scale at an earlier time under presumably diverse conditions. Additionally, situational factors in the daily life are way more varied than what can be contained or included in an experiment, so there might be other situational influence that might affect the process than the one put under examination.

    • Gregory Rosen says:

      Hi Benji, Regarding Study 2, do you think the problem with assuming another person’s feeling is due to the false consensus effect? What other types of bias might be at play, e.g., (actor-observer asymmetry/fundamental attribution error)? I appreciate your environmental perspective. What additional measurements from your field might be used for the other situational factors in daily life?

  9. Dakota Egglefield says:

    Despite the flaws mentioned in comments above, I liked the TASS model in relation to aggression, but I wonder how generalizable it is. I follow the threshold theory to a point, where a low aggressive person may need a higher situational threshold trigger to then show their aggressiveness, and that a person high in aggression needs a low situational threshold to trigger their aggression. However, when thinking about applicability to other traits, the second part of the theory, that people low in the trait will have a stronger reaction to a high-risk or provocative situation than people high in the trait, does not resonate with me. This brings me to think about how applicable it would be to other traits. In the case of altruism, it seems that the TASS model would be consistent only for the first part of the theory.

    Consider 3 situations: low – a person sitting on the street with no sign and not begging for money; moderate – a person on the train asking for donations; high – a person walking down the sidewalk, stopping, and beginning to seize. If we apply the TASS model to these situations using the trait of altruism, a person high in altruism may be more inclined to donate a dollar to a person on the train (moderate provocation) than a person low in altruism (let’s pretend we are in a city other than NYC, where the majority of individuals are not emotionally numb to the homeless population). However, according to the TASS model, a person low in altruism will have a stronger altruistic reaction towards a person seizing on the sidewalk than that of someone high in altruism. I am not a personality theorist not an altruism specialist and therefore somewhat naïve in this area, so my thoughts could be wrong, but I don’t buy this. I would expect a person high in altruism to react just as strongly, if not more strongly, than someone low in altruism. The effect of the low-altruism person’s reaction may seem more intense due to the fact that they would probably not respond much to a moderate or lowly provocative situation, but I believe that is based on an individual threshold and should not necessarily be compared to those high in altruism (or aggression, or any trait). I do recognize that altruism and aggression are two totally different traits dimensionally, but this is just an example of the lack of generalizability of the second facet of this model.

    As a few others touched on, the grouping of these three studies does not particularly add strength to the article although it seems like that was the author’s intent. While it is quite an interesting concept, I personally would have been more interested in reading an article applying the TASS model with the same study design to three different traits, even if they were traits highly similar and correlated with aggression instead of the aforementioned altruism.

    • Ariel Zucker says:

      Dakota – I definitely relate to your comment. It also did not resonate with me that the authors hypothesized that people low in the trait will have a stronger reaction to a strong provocative situation than people high in the trait. I feel that your example really highlights the problem with this hypothesis and provides a relatable perspective. The authors suggest that their model is most applicable to traits that have a biological basis, especially one’s relevant to emotionality (i.e., anxiety, self-esteem, etc). However, I feel that your altruism example can easily be applied to both anxiety and self-esteem and I would still align myself with your hypothesis rather than the authors’ hypothesis.

  10. Crystal says:

    One of the main findings of the article seems intuitive, that those with high levels of TA are more likely to express aggressiveness in mild-moderately anger-stimulating situations. To me, it makes sense that aggressive behavior would be generated more easily in someone who has a higher level of aggressiveness and anger generally. The finding that participants with low TA exhibited a more dramatic increase in aggressiveness in strong situations was a little surprising, at first. I think this is because those individuals with high TA are usually at a higher level of aggressiveness, so they can’t go much higher when the situation becomes stronger, as compared to those with low TA, who can become much more aggressive. I wonder why it appears that aggression actually decreased a bit from no situational provocation to moderate situational provocation for those with low TA (figure 2), even though the finding was not significant.

    I agree with the researchers regarding the applicability of the TASS model to other traits of biological basis. I would think that for a trait such as anxiousness, these findings would remain consistent. I can imagine that in lower anxiety-provoking situations (e.g., talking in class), those with higher baseline or trait anxiousness would become more anxious than those with lower trait anxiousness. However, if the situation were to become stronger (e.g., a dissertation defense), I could see that those will low anxiousness would increase in anxiety dramatically, comparatively. I also makes sense that individuals with low self-esteem would be more reactive to failure than those with high self-esteem.

    In my clinical work, I have seen these findings to be rather accurate. I once worked with a patient who came in to work on anger & aggressiveness. STAXI-2 assessment results showed elevated anger temperament (85th percentile) and outward expression (90th percentile). MMPI-2-RF results showed anger-proneness (behavioral-externalizing behavior related to aggression). It seemed that only moderately anger-provoking situations would ignite an aggressive response (e.g., getting cut-off while driving). In my experience with this patient, I cannot recall a strong situation to compare to, so I wonder how the patient would be responded in comparison to the moderate situation. I think the implications for real-life situations is important to consider.

  11. minjung park says:

    After I read this article, what I like the most interesting thing is about the model that explain traits represent different sensitivities to situational provocation. I think this article is connected to the previous article by Cooperman & Ickes (2009) we dealt with in some points. The first post that I wrote up was about “Extroversion = talkativeness; Introversion = Quiet.” Before I read the two articles, I was confused about my changeable personality traits depending on different situations like meeting my friends or strangers. Now I know that a person’s shyness would manifest itself more clearly when the person is interacting with strangers than when the person is interacting with friends depending on a sensitivity to the situational context. Besides, I was surprised by the results that the researchers found. I thought that high TA participants would be more active to not only middle and strong situational provocation but also no provocation than low TA participants. However, through the studies, I could know that low TA participants are more active to strong provocation, and high TA people will appear to be more reactive than low TA people to moderate provocation.
    However, I am curious about whether this study had considered the participants’ surrounding environment. This study lacks information about the participants’ background. I think such information can have a significant impact on the outcome. Furthermore, there are two things need more deep thoughts: Gender ratio and limited participants (101 UW women in study 1, and 29 men and 61 women of 90 UW undergraduates in the study 2). Those problems might make the present study unreliable. Ironically, the author already acknowledged this limitation and talked them in the limitation part. Nevertheless, I want to know why the author continued the studies? If the researchers expanded the participants more broadly with various participants and equal gender ratio, the results would have been different or changed. Therefore, those things should be considered.

  12. Ariel Zucker says:

    While I am impressed by the author’s rigorous study design and thorough explanation of their methodological logic, I wonder if there’s an important confounding variable that might have been overlooked. Namely, impulsivity. The APA Dictionary of Psychology defines impulsivity as a “tendency to act on a whim, displaying behavior characterized by little or no forethought, reflection or consideration of the consequences” (VandenBos, 2007). Just because someone might have the tendency to be more aggressive (Trait Aggression – TA), does not mean that they will act on those aggressive tendencies. If impulsivity were to be taken into account (Trait Impulsivity – TI), I wonder if TI will moderate the effects of TA.

    Let’s take, for example, Person A (High TA, High TI), Person B (High TA, Low TI), Person C (Low TA, High TI), and Person D (Low TA, Low TI). Contrary to the TASS model, I hypothesize that Person A will be the only person to show heightened aggression when experiencing moderate situational provocation. The TASS model, in contrast, might predict that both Person A and Person B will show heightened aggression when experiencing moderate situational provocation because they both have High TA. I also wonder about the possibility that Person C might also show heightened aggression when experiencing moderate situational provocation. Even though Person C has Low TA, would their High TI be enough to cause them to show even a mild aggression?

    In a perfect world, research would take into account every possible confounding variable. I am not suggesting such an unrealistic method of research. I am, however, suggesting that when designing trait studies, it is important to consider what other traits might moderate the effects of the trait in study. To me, it seems like impulsivity might be one of those additional traits that should have been taken into consideration.

    VandenBos, G. R. (Ed.). (2007). APA Dictionary of Psychology. Washington, DC, US: American Psychological Association.

    • Sergiu Barcaci says:

      Hi Ariel, that’s actually very insightful! While this was a pretty good preliminary study, future studies should definitely build upon it by including traits such as impulsivity. These could strengthen previous findings or completely turn them on their head and lead us to learn newer things about how different traits interact with each other and skew results. Impulsivity would be a negative trait that would affect TA, but what about some positive traits that would prevent someone with High TA from lashing out? Like disciple for example.

    • drcb says:

      As it’s thought of in the social psych literature, there is a direct connection between aggression scores and aggressive acts (whether verbal, physical, etc.,). Impulsivity could further exacerbate that tendency though.

  13. Shira Russell-Giller says:

    I found Marshall and Brown’s (2006) Traits as Situational Sensitivies (TASS) Model to be an interesting way of conceptualizing personality traits, and I think that overall, their battery of studies supported their model of traits functioning as threshold reactions to degrees of situational provocation. Given the subjective nature of “situational provocation,” I appreciated the authors’ initial pilot study to verify that the feedback statements they would present in Study 1would in fact elicit different levels of provocation. Indeed, Study 1 supported their hypothesis that participants with high TA would respond more aggressively to moderate provocation than participants with low TA. However, I am left wondering how ‘moderate TA participants’ would have responded to moderate provocation. I understand that the authors needed to exclude moderate TA participants in order to demonstrate their hypothesis clearly, but the downside to this exclusion if that 1/3 of the original population is left in question. According the authors, the TASS model explains that when situational provocation reaches just above the threshold necessary to elicit trait-related reactions (i.e., moderate provocation), then those high in in the trait will respond more strongly than those low in the trait. I am not sure how the TASS model would predict the reaction of individuals with a moderate trait, but perhaps one possibility is to view the trait on a continuum rather than as low versus high. By viewing traits on a continuum, where every individual falls somewhere on the continuum from lowest to highest, we can match each point on the continuum to a specific situational-provocation threshold that is necessary to elicit a trait-relevant reaction. Hypothetically, once a specific provocation threshold is paired with each point on the trait continuum, we can follow the incline or decline of the trait reaction as the situational provocation increases to determine the pattern for individuals with a moderate level trait.

    • Shanna Razak says:

      I agree I believe that is why the researchers left out testing the middle part of the aggressiveness spectrum. When it does come to that portion of the spectrum, I do not believe that a person can be precisely in the middle. Although one may not be in the extreme ends of the trait aggressiveness spectrum they will either be slightly more or less than the median. Therefore I believe that they would comply with the TASS model in that their individual differences will show in moderate provocation and be eliminated in strong provocation

  14. Ulzhan says:

    The main findings of the Article at first seemed very logical to me that people with high TA would react more aggressive towards moderate and high sensitivity situations, compare to low TA participants. However, after reading more into the study and analysis presented by the author made me wonder about few confounds that may explain why people with high TA would not show a difference in their response when compared with moderate and high sensitivity situations. People with high TA might have shown even higher aggressiveness towards the situation if the scale allowed compared to people with low TA at the high sensitivity situation. After analyzing the results of the Study 2 and 3, I was surprised to find out that participants’ prediction showed that even people with low TA showed some reaction towards moderate sensitivity situation whereas the Study 1 did not. This could be an interesting discussion of how the situational sensitivity defers in people who personally experience or giving a hypothetical judgment of the other person or self.
    I would like to note, however, that I agree with the author that the results during moderate sensitivity situations could be different due to personality effects. I believe that it is more interesting to analyze than comparing how people with low and high TA would respond to different situations that provoke aggression. It looks like the study concentrates on the environmental attributes that could affect the responsiveness of the person to a situation.
    It was interesting to know that most of the previous studies were compared lows and highs in TA and the sensitivity situation, and this study indicated an existing threshold people have that can be analyzed further. I believe we all have a certain threshold when assessing the situations that can provoke negative feelings/aggression. Ultimately, I think it is safe to say that in ambiguous situations we tend to rely on experience, standards, and our personality.

  15. Shanna Razak says:

    Traits are situational based, in that the behaviors regarding the traits are exhibited based on the environmental surroundings. Sensitivity to a situation depends on how high of a trait one possesses(at least up to the moderate provocation). When reaching strong provocation those that are on the low and high ends of the trait spectrum will show behaviors of that trait. For example in the “strong provoked” situation of losing one’s job,those who are neurotic on any part of the spectrum will exhibit neurotic behaviors
    The study itself had many limitations. One limitation that stood out to me was that fact that an indirect measure of aggression was used. Also this act was done anonymously (the person meant to be punished was not present). If direct infliction of aggression was used it may be much more accurate. However, I believe that there is a possibility that it would have underestimated the aggression of the participant If the participant actually met the person they are “punishing” they may feel guilty for wanting to administer a very loud sound causing them to choose a softer sound. This would not accurately reflect their aggressiveness.
    This can be connected to Milgram’s study in which deception and mental harm of the participants was a huge concern. The use of a direct aggression approach would result in a similar yet not as extreme issue found in the Milgram study. The reason I say less extreme version is that the guilt that may be attributed to “administering” a loud noise is nothing compared to “discharging” a deathly shock. Hypothetically after being informed that the noise was not delivered one may be shocked at the effect of their aggression and how it controls their actions( similar to the Milgram’s participant realizing how easily they could be manipulated into hurting another person)

    • Bengi Sullu says:

      Hi Shanna,
      I was also left with the question of the implications of not having the person that they were supposed to show aggression against. I was wondering about ways in which aggression can be measured in the context of an interaction within daily life, something that is already taking place, especially after reading your reference to Milgram’a study. That even applies to the anger part of the measurement as even then it was no direct contact.

    • drcb says:

      There are certainly ethical issues in the measurement of aggression in research. Typically, noise bursts or hot pepper shots (amount administered) are used as stand-ins for actual physical aggression. Study 1 was unusual in that the subjects didn’t directly “administer” the noise.

  16. Gregory Rosen says:

    To start, the Allport quote about perspiration and shivering confused me, since people perspire for reasons other than heat, such as nervousness, and shiver for reasons other than cold, such as fear. Moreover, heat and cold do not seem like helpful analogies for social provocations involving higher order cognitions. Sweating and shivering due to temperature are often automatic responses, while the emotions and behaviors described in these studies involve more complex mental processing.
    What most stood out to me overall was the opinion regarding Study 2, that like lay perceivers in general, the study participants were good at predicting others’ behavior (implicit TASS theorists), versus the opinion regarding Study 3, that individuals generally aren’t always able to accurately predict their own emotional responses. The results of Study 3 indeed showed that both low TA and high TA groups expected to experience more anger due to more provocation, which is not always accurate according to TASS. If we accept these results and the given generalizations, what does this say about evolutionary advantages? Why would people be better at predicting others’ behavior than their own? Allow me to speculate, but is it possible that individuals sacrifice prediction-error-minimizing capacity toward themselves in order to more smoothly co-exist within their group and contribute to its success? Are we wired for social living at the expense of becoming more delusional about ourselves? The study mentions self-esteem at the end. I wonder if both high and low self-esteem are different aspects of the same illusion – the illusion that there is a unified self at all. Self-esteem could be driven by a sense of how meaningful one’s life feels, in terms of how much one contributes to the group. In this way, the illusions of selfhood, self-esteem, meaningfulness, and importance of social contribution serve the group while keeping each individual believing in false constructs about themselves.
    Finally I was confused why the authors hinted that some traits have a biological basis, while others may not. How do they determine what traits do not have a biological basis? Is this a philosophical mind-brain dualist stance?

  17. Gregory Rosen says:

    Hi Benji, Regarding Study 2, do you think the problem with assuming another person’s feeling is due to the false consensus effect? What other types of bias might be at play, e.g., (actor-observer asymmetry/fundamental attribution error)? I appreciate your environmental perspective. What additional measurements from your field might be used for the other situational factors in daily life?

  18. Gregory Rosen says:

    My comment landed in the wrong place, so I’ll re-submit it above under Benji’s post.

Leave a Reply




Spam prevention powered by Akismet

Skip to toolbar