The Neurobiology of Borderline Personality Disorder

Are there any structures or biological abnormalities that make one more likely to have a borderline personality disorder?

Prediction: I can imagine that an amygdala abnormality or perhaps dopamine receptor deficiency similar to those with addiction. I hesitate to rely on a biological explanation due to the social developmental issues those with borderline personality disorder commonly have such as a non-cogent family structure.

Article 1:

Borwin Bandelow, Christian Schmahl, Peter Falkai, Dirk Wedekind, Borderline Personality Disorder: A Dysregulation of the Endogenous Opioid System?, Psychological Review, Volume 117, Issue 2, April 2010, Pages 623-636, ISSN 0033-295X, 10.1037/a0018095.

(http://www.sciencedirect.com/science/article/pii/S0033295X10600527)

Validity: This journal is a registered trademark of the APA. APA has very high standards of validity. The association states “[it] is a violation of APA Ethical Principles to publish ‘as original data, data that have been previously published’ (Standard 8.13).

In addition, APA Ethical Principles specify that ‘after research results are published, psychologists do not withhold the data on which their conclusions are based from other competent professionals who seek to verify the substantive claims through reanalysis and who intend to use such data only for that purpose, provided that the confidentiality of the participants can be protected and unless legal rights concerning proprietary data preclude their release’ (Standard 8.14).

APA expects authors to adhere to these standards. Specifically, APA expects authors to have their data available throughout the editorial review process and for at least 5 years after the date of publication.

Authors are required to state in writing that they have complied with APA ethical standards in the treatment of their sample, human or animal, or to describe the details of treatment.” No further information was available at this level of access.

Summary:

This study studied several possible neurobiological dysfunctions such as serotoninergic, dopaminergic, and other neurotransmitter systems. The most promising was the opiod receptors. They questioned the sensitivity of the receptors and availability of the endogenuous opiods. The unhealthy behaviors that are symptomatic to BPD possibly occur stimulate the opiod system and dopaminergic reward system. The study explains clearly: “Frantic efforts to avoid abandonment, frequent and risky sexual contacts, and attention-seeking behavior may be explained by attempts to make use of the rewarding effects of human attachment mediated by the EOS.” The reduced activity of the same are may be causative of anhedonia (the inability to feel pleasure) and the feelings of emptiness. The substance abuse also damages those receptors. Other control oriented behaviors like food behaviors like restriction, self-mutilation, and aggressive behavior is to set the body in survival mode which depletes the last reserves of the EOS. Opiod receptor antagonists can control many of these behaviors.

Reflections and thoughts: I was pleased that the answer was relatively clear that by addressing the endogenous opiod system that borderline personality disorder could be that closer to a cure pharmacologically. It is interesting that these behaviors actually are falsely placing the body in a state of “fight or flight” and that feels like relief to those with BPD. It also explains the popularity of heroin use I have personally seen in those with BPD

Article 2:

Joel Paris, Hallie Zweig-Frank, N.M.K Ng Ying Kin, George Schwartz, Howard Steiger, N.P.V Nair, Neurobiological correlates of diagnosis and underlying traits in patients with borderline personality disorder compared with normal controls, Psychiatry Research, Volume 121, Issue 3, 1 January 2004, Pages 239-252, ISSN 0165-1781, 10.1016/S0165-1781(03)00237-3.

(http://www.sciencedirect.com/science/article/pii/S0165178103002373)

Validity: Although the article is seven years old, I am comfortable with the validity. The journal has high standards of validity stating that “[submitted] manuscripts will be reviewed anonymously by at least two referees. Should a revised manuscript be required by the editors, the authors are requested to resubmit their revised manuscript to the journal within 6 months time. Studies on humans submitted to the journal must comply with the principles laid down in the Declaration of Helsinki (Br Med J 1964; 2: 177-178). The editors retain the right to reject papers on the grounds that, in their opinion, the ethical justification is questionable.”

The small sample size is arguably less valid. The large number of controls is creates more validity as well as the barrage of tests. More results in general gives a more complete picture. No further information was available at this level of access.

Summary:

This study was also focused on neurotransmitter systems. Thirty women with the disorder plus twenty two controls went through the following tests: Diagnostic Interview for Borderlines, revised, the Hamilton Depression Scale (HAM-A) and the Hamilton Anxiety Scale (HAM-A), the Diagnostic Assessment of Personality Pathology, the Buss–Durkee Guilt-Hostility Inventory, the Barratt Impulsivity Scale (BIS), and other tests to measure serotonergic, cholinergic and noradrenergic activity. Results point to issues with sertogenic systems, that are related to serotonin, as most pertinent to the disorder.

Reflections: Most of those with BPD that I know have been in dual diagnosis wards. They are heavily medicated with SSRIs. This closely relates to this study. It also made me realize that as far as validity the large amount of tests made me feel more confident in the study. Interesting that an increase in data can be more convincing. My ignorance of the actual test methods lead me to believe that this may not make the study more valid.

Article 3:

Tsutomu Takahashi, Andrew M. Chanen, Stephen J. Wood, Mark Walterfang, Ian H. Harding, Murat Yücel, Kazue Nakamura, Patrick D. McGorry, Michio Suzuki, Dennis Velakoulis, Christos Pantelis, Midline brain structures in teenagers with first-presentation borderline personality disorder, Progress in Neuro-Psychopharmacology and Biological Psychiatry, Volume 33, Issue 5, 1 August 2009, Pages 842-846, ISSN 0278-5846, 10.1016/j.pnpbp.2009.03.035.

(http://www.sciencedirect.com/science/article/pii/S0278584609001080)

Validity: This journal again had high standards for validity. It only accepts original research or mini reviews of work done in the last two years. The work must current and relevant. Similar statements on disclosure of conflicts of interests and ethics are enforced. All work must not have been previously published and be original. Ethics must be practiced: “The Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving humans; EU Directive 2010/63/EU for animal experiments; Uniform Requirements for manuscripts submitted to Biomedical journals. This must be stated at an appropriate point in the article.” No further information was available at this level of access.

Summary:

This studied differed from the previous studies in that it focused on potential midline brain abnormalities that may be found in those with BPD. The study included 20 in the sample and 20 controls. They used an MRI to study the length of two structures adhesio interthalamica (AI) and cavum septum pellucidum (CSP) and the volume of the third ventricle. as well as third ventricular volume in 20 teenagers with first-presentation BPD and 20 healthy controls. The findings were that “the AI was significantly shorter in BPD patients than in controls. Furthermore, the BPD patients had a significantly larger third ventricle than controls.” The conclusion of the researchers was that further neuroimaging can help.

Reflections: Strangely what strikes me most about this study is the frustration with the conclusion. All that they were comfortable with saying was that?! It was such a let down. It also reminded me of the brain size studies that compared the sexes. How unfulfilling!

Article 4:

Katja Wingenfeld, Carsten Spitzer, Nina Rullkötter, Bernd Löwe, Borderline personality disorder: Hypothalamus pituitary adrenal axis and findings from neuroimaging studies, Psychoneuroendocrinology, Volume 35, Issue 1, January 2010, Pages 154-170, ISSN 0306-4530, 10.1016/j.psyneuen.2009.09.014.

(http://www.sciencedirect.com/science/article/pii/S0306453009002911)

Validity: The recent date of the publication was a promising sign of validity. The journal also has an international reputation. Articles are mostly acquired by invitation only. When submitted, policies for submission are strict requiring 4-5 additional reviewers with submission, full disclosure of potential or actual conflicts of interest, and a description of general details of support from sources for the study. The journal warns “Failure to comply with conflict of interest requirements may result in rejection of the manuscript and/or retraction of the published article. Disclosure of the conflict of interest will be made in the published article. There is no disclosure form as such, you only need to mention any conflict of interest by mentioning its nature, but only when there is one.” No further information was available at this level of access.

Summary:

The focus of this study on was the hypothalamic-pituitary-adrenal (HPA) axis in BPD, compared to others such as bipolar and PTSD. They used neuroimaging to assess the stress that may damage possibly the hippocampus. Results pointed to incressed suppression in the HPA Axis.

Reflections: I did like that this study compared other disorders neurobiology. The symptoms of bipolar and PTSD are very similar to BPD. Thusly I was surprised that BPD individuals had increased suppression. In a lot of ways BPD seems safer. Maybe increases options/solutions for those with BPD made me on some level think that it would not show a difference. I would be interested in seeing how MRI’s might detect improvement in those with BPD who had used DBT.

The Brain and PTSD

Amber Badgerow

November 15, 2011

PSY 213

John Olmsted

The Brain and PTSD

Question:

Post-traumatic stress disorder (PTSD) is an anxiety disorder that arises from a traumatic event. The effects can be devastating to a person’s life. One of the main issues in PTSD is that a patient is unable to suppress fear even in safe conditions. Also, patients have more vivid memories to the point of reliving the experience. What occurs in the brain that may be associated with the overactive fear and memory response in PTSD?

Prediction:

A person with PTSD may exhibit a hyperactive amygdala, hippocampus, and frontal cortex. These areas are involved with emotions like fear and the formation and storage of emotional memories.

Article 1: “Amygdala activity correlates with attentional bias in PTSD”

This article discusses the study conducted by Myriam El Khoury-Malhame, Emmanuelle Reynaud, Alice Soriano, Keller Michael, Pilar Salgado-Pineda, Xavier Zendjidjian, Caroline Gellato, Fakra Eric, Marie-Noelle Lefebvre, Frank Rouby, Jean-Claude Samuelian, Jean-Luc Anton, Olivier Blin, and Stephanie Khalfa. The study was designed to examine the relationship between the activity of the amygdala, the intensity of PTSD symptoms, and attentional bias in patients with PTSD. During fMRI scanning, seventeen patients with PTSD and seventeen healthy subjects performed an attentional detection of target (DOT) task- where they concentrated on a cross on a screen while quickly locating a dot elsewhere- and a matching task- where they identified whether or not an emotional facial expression or a neutral geometric shape was the same as another. The patients with PTSD exhibited increased amygdala activity when presented with emotional faces and the extent of the activity related to the severity of the symptoms and the degree of attentional bias. Greater attentional bias was also associated with increased activity in the hippocampus and prefrontal cortex. Their results provided additional evidence of hyperactivity in the amygdala in patients with PTSD. They suggested that the increased activity in the amygdala is related to heightened perceptual responsiveness to threats.

Article 2: “Abnormal baseline brain activity in posttraumatic stress disorder: A resting-state functional magnetic resonance imaging study”

Yan Yin, Lingjiang Li, Changfeng Jin, Xiaolei Hu, Lian Duan, Lisa T. Eyler, Qiyong Gong,

Ming Song, Tianzi Jiang, Mei Liao, Yan Zhang, and Weihui Li conducted a study to examine regional baseline brain function in patients with PTSD using fMRI to obtain an amplitude of low-frequency fluctuation (ALFF) measure. They explained that results from previous studies of abnormal brain activity and PTSD differed and that a resting state, without cognitive tasks, would offer more accurate results since it would eliminate more variables. They requested fifty-four patients with PTSD and seventy-two comparison subjects to rest comfortably and not think about anything in particular while having their brains scanned. The team found that there was lower ALFF values in the visual cortex, cerebellum, and right insula of the patients with PTSD compared to the control group. They also exhibited increased ALFF values in the dorsolateral prefrontal gyrus and the ventral medial frontal gyrus. Greater ALFF values correlated with more severe symptoms in patients with PTSD. The decrease in insular and cerebellar activity may be related to inadequacies in memory. Decreased visual cortex activity could be associated with difficulties in verbal memory and visual imagery. The increased activity in the medial prefrontal cortex may have a correlation to mental re-experiencing of traumatic events. Greater activity in the dorsolateral prefrontal cortex could be associated with the vivid recollection of traumatic events that patients with PTSD experience.

Source Validity:

The articles were written by reliable and credible sources. They both presented thorough examples from previous research that added to the accuracy of the results of their study. Despite having a small sample group, both sources controlled well for age and education and had specific criteria for establishing a correct PTSD diagnosis with no history of other mental conditions. The second study further ensured that all patients with PTSD were recent victims and had not been on medication. The second study, however, could have had interference in results from some medicated patients, although they did try to eliminate that possibility with extensive analysis of the data.

Reflection on Hypothesis:

The articles found in this research support my hypothesis that people with PTSD have hyperactivity in the amygdala and frontal cortex. Although the first article mentioned activity in the hippocampus, it was brief and there wasn’t enough specificity to state that it supported my hypothesis. The research added to my prediction in the finding of decreased activity in certain brain regions that may be related to other symptoms of PTSD not focused on in my question.

Works Cited:

Khoury-Malhame Myriam El, Reynaud Emmanuelle, et al. “Amygdala activity correlates with

attentional bias in PTSD.” Neuropsychologia 49.7 (2011): 1969-1973. Elsevier Ltd. ScienceDirect. Web. Nov. 14, 2011.

http://0-www.sciencedirect.com.library.pcc.edu/science/article/pii/S0028393211001709

Yin Yan, Li Lingjiang, et al. “Abnormal baseline brain activity in posttraumatic stress disorder:

A resting-state functional magnetic resonance imaging study.” Neuroscience Letters 498.3 (2011): 185-189. Elsevier Ireland Ltd. ScienceDirect. Web. Nov. 15, 2011.

http://0-www.sciencedirect.com.library.pcc.edu/science/article/pii/S0304394011002606

The Emotional Pathways of Bipolar Disorder

Amber Badgerow

November 13, 2011

PSY 213

John Olmsted

The Emotional Pathways of Bipolar Disorder

Question: Bipolar is a tragic, debilitating disorder that is distinguished by the fluctuation between emotional extremes of depression and mania, in the case of Bipolar I Disorder, or hypomania in the case of Bipolar II Disorder. Is there an abnormality of the brain’s emotional pathway, specifically in the amygdala and prefrontal cortex, in those with Bipolar Disorder?

Prediction: The brain’s emotional pathway in those with Bipolar Disorder may be hyperactive in the in amygdala, or may have irregular connections, or may be hypoactive in certain regions, specifically the prefrontal cortex that cognitively regulates emotion.

Article 1: “Functional Magnetic Resonance Imaging Brain Activation in Bipolar Mania: Evidence for Disruption of the Ventrolateral Prefrontal-Amygdala Emotional Pathway”

This article discusses the study by Stephen M. Strakowski, James C. Eliassen, Martine Lamy, Michael A. Cerullo, Jane B. Allendorfer, Michelle Madore, Jing-Huei Lee, Jeffrey A. Welge, Melissa P. DelBello, David E. Fleck, and Caleb M. Adler to detect irregularities in the emotional system of patients with manic bipolar using functional magnetic resonance imaging (fMRI). Previous research indicated that a ventrolateral prefrontal network regulated external emotional stimulus processing and Strakowski et al. chose to study this network. Forty patients with bipolar mania and thirty-eight healthy subjects performed a Continuous Performance Task with Emotional and Neutral Distracters (CPT-END) while brain scans were taken. Patients with bipolar mania exhibited dulled activation to emotional and neutral stimulus in the right lingual gyrus, the medial thalamus, the left ventrolateral prefrontal cortex, the left amygdala, the right parahippocampus, and the precuneus compared to the control group. However, they exhibited hyperactivity to "targets"- colored circles that required participants to press a button different from the one pressed for emotional or neutral pictures- in the right inferior frontal gyrus, the left fusiform gyrus, the right superior frontal gyrus, the right parahippocampus, the right putamen, and the right amygdala. This suggested that patients with bipolar enlisted that help of other brain regions for the task more than the healthy subjects did. These results indicated that the ventrolateral prefrontal-amygdala emotional network does not function properly during bipolar mania and causes a loss of regulation of the limbic brain by the ventral prefrontal cortex.

Article 2: “Abnormally increased effective connectivity between parahippocampal gyrus and ventromedial prefrontal regions during emotion labeling in bipolar disorder”

This article was written by Jorge R.C. Almeida, Andrea Mechelli, Stefanie Hassel, Amelia Versace, David J. Kupfer, and Mary L. Phillips. It reviews their study to examine remaining irregularities in the effective connectivity of the ventromedial and dorsal/lateral neural systems associated with mood modulation in patients with subsiding bipolar disorder. They used dynamic causal modeling (DCM) and fMRI to determine effective connectivity while participants (twenty-one patients with bipolar disorder and twenty-five healthy subjects) viewed mild and intense happy and neutral facial emotions. Patients with bipolar disorder exhibited reduced activity in the right parahippocampal gyrus with intense and mild happy faces compared to the healthy control group. They also showed decreased activity in the right dorsolateral prefrontal cortex in response to all emotional faces. Greater effective connectivity was seen in patients with bipolar between the right parahippocampal gyrus, which has connections to the amygdala, and the right subgenual anterior cingulate gyrus, associated with depression, when identifying emotions of happy and neutral facial expressions. These results indicated that patients with bipolar disorder have a dysfunctional ventromedial prefrontal network- associated with the early evaluation of stimulus, heightened encoding, and the automatic modulation of emotion- which may be a cause of the mood dysregulation observed in bipolar disorder.

Source Validity:

Both sources had small sample groups and had a mix of medicated and non-medicated patients with bipolar. The first study only had patients that were manic and the second had only had patients that were in remission. However, they also controlled for age, race, IQ, education, handedness, and other medical conditions, specifically psychological. Also, both sources were reliable and referenced previous studies to support their results. The second article was a little older, but still presented results that are concurrent with most other studies. One of the factors that lessened the reliability of the studies mentioned above was the lack of repeated studies that used the same task to find abnormalities in the emotional system of patients with bipolar disorder.

Reflection on Hypothesis:

The research supported my hypothesis that there may be irregular connectivity in the emotional system in people with bipolar disorder. However, increased and decreased activity in certain brain regions that are involved in processing and regulating emotion depends on the task participants perform. In some studies not mentioned in this article the amygdala showed increased activity, but in the first article discussed there was decreased activity. The research did support my prediction that the prefrontal cortex may show decreased activity, but more studies need to be reviewed to ensure validity. Many other studies have examined emotional pathways in euthymic bipolar, but the above articles were chosen because I wanted to concentrate on people with manic bipolar.

Works Cited:

Strakowski Stephen M., Eliassen James C., et al. “Functional Magnetic Resonance Imaging

Brain Activation in Bipolar Mania: Evidence for Disruption of the Ventrolateral Prefrontal

Amygdala Emotional Pathway.” Biological Psychiatry 69.4 (2011): 381-388. Elsevier Inc.

ScienceDirect. Web. Nov. 13, 2011.

http://0-www.sciencedirect.com.library.pcc.edu/science/article/pii/S000632231000956X

Almeida Jorge R.C., Mechelli Andrea, Hassel Stefanie, Versace Amelia, Kupfer David

J., and Phillips Mary L. “Abnormally increased effective connectivity between

parahippocampal gyrus and ventromedial prefrontal regions during emotion labeling

in bipolar disorder.” Psychiatry Research: Neuroimaging 174.3 (2009): 195-201.

Elsevier Ireland Ltd. ScienceDirect. Web. Nov. 13, 2011.

< http://0-www.sciencedirect.com.library.pcc.edu/science/article/pii/S0925492709001176>

The Phenonmenal Production of Dreams

Amber Badgerow

November 7, 2011

PSY 213

John Olmsted

The Phenomenal Production of Dreams

Question:

Dreaming is an amazing phenomenon. It involves vivid visual, kinesthetic, and audio mental imagery that occurs while one is asleep and has only a partial connection to environmental stimuli. The function of dreaming is in much debate and a popular topic of research, but not as much research has been devoted to the neural mechanisms of dreaming. How are dreams, that incorporate mental sensory imagery, emotional experiences, and cognitive processes, generated?

Prediction:

Dreams may involve similar networks associated with daydreaming/mind wandering and the mind’s eye. The production of dreams may be a more or less whole brain process that incorporates the Parieto-Occipito-Temporal junction (PTO), which is responsible for perceptual processing, the frontal lobes of the brain, and the limbic system. This would connect sensory, memory, thought process, and emotions to dreaming.

Article 1: “Cognitive and emotional processes during dreaming: A neuroimaging view”

This article was written by Martin Desseilles, Thien Thanh Dang-Vu, Virginie Sterpenich, and Sophie Schwartz to collectively review neuroimaging studies on rapid eye movement (REM) sleep that provide insight into the neurobiology of dreams. REM sleep is associated with more vivid dreams and more occurrences of dreams compared to non-rapid eye movement (NREM) sleep. Neuroimaging of the brain has showed that the basal forebrain, the pontine tegmentum, the limbic/paralimbic region, and the thalamus have increased brain activity during REM sleep. Studies have also found that the basal ganglia, cerebellum, and the primary motor and premotor cortices (all motor regions) are activated during REM sleep, which could relate to the movement element in dreams. It has also been found that, while the primary visual cortex is inactivated during REM sleep, the visual association regions show significant activation. This could indicate that mental visual information is processed in a network separate from that of external stimuli processing, which would explain how those with blindness caused by brain damage can still have visual dreams. Activity was seen in the precuneus, suggesting a correlation between dreaming and the default mode network (DMN), which has been proposed to form a central imagery network through high-level cognitive processes that promote mental imagery. Furthermore, activation of the temporal and occipital cortex during REM sleep might be the underlying mechanism of auditory and visual content in dreams. During REM sleep, decreased activity in the parietal cortex and the lateral and inferior prefrontal cortex could explain why people easily accept bizarre elements of dreams and exercise little control over dreams' proceedings. Activation of the medial prefrontal cortex may contribute to a dreamer's ability to associate emotions, intentions, and thoughts to people in dreams. These studies give an idea of the neural mechanisms that produce the sensory, emotional, and cognitive elements of dreams.

Article 2: “The neural substrate for dreaming: Is it a subsystem of the default network?”

G. William Domhoff wrote this article to present studies that support the idea that the neural mechanism of dreaming is related to the default network. Neuroimaging has shown that the default network is active during simulation, mind wandering, and daydreaming. Studies have also shown activity of this network during the onset of sleep and some activity during REM sleep. Lesion studies, that associated the anterior cingulate cortex, medial prefrontal cortex, basal forebrain, and the limbic region with dreaming, were supported by neuroimaging studies that found that those same regions were more active during REM sleep than NREM.

The auditory and visual association cortices, the pontine tegmentum, and the occipital-temporal lobe have also been found to be active during REM sleep. More recent studies show that the dorsolateral prefrontal cortex is more active in REM than NREM, which differed from earlier studies that found the region to be less active during both stages of sleep. The neural mechanisms of dreams have a connection to the default network because they both promote unregulated thought that is independent of stimulus. Research by Smallwood and Schooler provide similarities between mind wandering and dreaming. Simulation can be involved in mind wandering as in dreaming, but also involves self-reflective unsystematic thinking. Functional MRI studies of mind wandering further support the connection since many of the brain regions activated during sleep/dreaming were also active during mind wandering. A study by Christoff et al. (2009) identified the two fundamental governing brain areas of the default network- the dorsolateral prefrontal cortex and the dorsal anterior cingulate cortex. Subsystems of this network were suggested by further studies. However, there is evidence that the posterior regions of the default network are not very active in REM sleep, which has caused scientists to doubt the possibility of the dreaming network as a subsystem of the default network. Yet, still others reason that the dream subsystem has no need for external monitoring or episodic memories and therefore may be able to function without those areas. The similarities cannot be ignored and only further research can help determine whether the workings of dreaming are a subsystem of the default network.

Source Validity:

Although both sources were from credible authors, they were collective reviews of numerous studies, which has both positive and negative aspects toward validity. Collective reviews provide

numerous sources of specific information that can be seen as a whole. Also, they present repeated studies to ensure accurate results. However, it is difficult to examine the methods, subjects, and particular goals of each study referenced. Therefore, the validity/reliability of each source in a collective review is somewhat unknown. The author of a collective review is also able to extract only certain information from studies for his or her purposes. Since there was a single author, the second article is more questionable as bias could be a greater issue. Despite these issues both articles presented useful information to understanding the neurobiology of dreams.

Reflection on Hypothesis:

The research supported my hypothesis that the neural mechanisms of dreaming involve the PTO, frontal lobes, and limbic system and may be related to the default network. It also revealed that the visual and auditory cortices were active during REM sleep, which is associated with dreaming.

Works Cited:

Desseilles Martin, Dang-Vu Thien Thanh, Sterpenich Virginie, and Schwartz Sophie. “Cognitive

and emotional processes during dreaming: A neuroimaging view.” Consciousness and

Cognition 20.4 (2011): 998–1008. Elsevier Inc. ScienceDirect. Web. November 7, 2011.

http://0-www.sciencedirect.com.library.pcc.edu/science/article/pii/S1053810010001935

Domhoff G. William. “The neural substrate for dreaming: Is it a subsystem of the default

network?” Consciousness and Cognition 20.4 (2011): 1163–1174. Elsevier Inc.

ScienceDirect. Web. November 7, 2011.

http://0-www.sciencedirect.com.library.pcc.edu/science/article/pii/S1053810011000468