Friday, January 27, 2012


Today was the sixth anniversary of this blog. I'm not much for meta-blogging or general chattiness, but I thought I would highlight the nine posts (out of 700) with the most comments. Thank you for your support over the years, and keep the comments coming.

9. Friston Is Freudian - Friday, March 12, 2010

Neuropsychoanalysis is in the news again because of the recent publication of Neural correlates of the psychedelic state as determined by fMRI studies with psilocybin. In 2010, first author Carhart-Harris published an expansionist mega-opus (with Karl Friston) on The default-mode, ego-functions and free-energy: a neurobiological account of Freudian ideas, the basis for the present post and its follow-up.

8. Is CBT Worthless? - July 03, 2009

According to a meta-analysis by Lynch, Laws and McKenna, Cognitive Behavioral Therapy (CBT) is not helpful for those with schizophrenia and bipolar disorder, and any improvements seen in major depression are rather small.

7. White Matter Differences in Pre-Op Transsexuals Should NOT be the Basis for Childhood Interventions - January 28, 2011

Contains a number of comments by transgendered individuals who took exception with various aspects of this post.

6. The Precuneus and Recovery from a Minimally Conscious State - July 05, 2006

Includes a number of comments, over a two year period, from a father caring for his son.

5. Voodoo Correlations in Social Neuroscience - January 05, 2009

On the infamous paper by Edward Vul, Christine Harris, Piotr Winkielman and Harold Pashler, ultimately retitled Puzzlingly High Correlations in fMRI Studies of Emotion, Personality, and Social Cognition.

4. Glossolalia - November 04, 2006

Includes personal statements from many individuals who feel they speak in tongues.

3. Bad News for the Genetics of Personality - August 07, 2010

A recent search for genetic variants that underlie differences in personality traits came up empty (Verweij et al., 2010).

2. The Pseudoscience of Anti-Psychiatry in PLoS Medicine - August 01, 2006

Antipsychiatry is always a hot-button topic, and this early post attracted 44 comments.

1. Airplane Headache - August 15, 2010

The winner by a mile, with 72 comments, is on a supposedly rare type of headache that occurs during take-off and landing (Atkonson & Lee, 2004). The pain appears to be unique to plane travel and not associated with other conditions. Neurological exam and brain imaging results in all published cases (n=14) have been normal. Clearly, there are more than 14 people who suffer from these excruciating headaches on airplanes. Triptan drugs (used to treat migraines and cluster headaches) may be effective in preventing airplane headaches (Ipekdal et al., 2011).

Thank you for reading!

Subscribe to Post Comments [Atom]

Thursday, January 19, 2012

Deep Brain Stimulation for Bipolar Depression

The Melancholia of Kirsten Dunst and Lars von Trier

“Gray wool, clinging to my legs, it's heavy to carry along”

The disastrous wedding reception of the severely depressed Justine precedes the end of the world, depicted as a highly stylized and artistic event feared by some but welcomed by others. Kirsten Dunst plays the role of von Trier's own melancholia, which was the inspiration for his film.

The image above occurred out of context, at the very beginning, during the bombastic Wagnerian apocalyptic prelude to Part One, "Justine" and Part 2, "Claire." We don't hear Justine say those words until later, when she had lost the ability to care for herself. "She should be hospitalized," I thought at the time, and wondered why no one was getting her psychiatric help. But then we wouldn't have a movie that deals with internal struggle and suffering.

Deep Brain Stimulation for Treatment-Resistant Depression

Severe depression that is refractory to treatment, i.e. unresponsive to psychotherapy, multiple trials of antidepressant drugs (often combined with atypical antipsychotics, mood stabilizers, benzodiazepines, etc.) and electroconvulsive therapy (ECT), takes a tremendous toll on the long-suffering patients and their families. An alternative treatment modality, deep brain stimulation (DBS), has been in clinical trials for intractable depression for nearly 10 yrs. It works using the same sort of device used in DBS for Parkinson's disease, which has been remarkably successful in alleviating symptoms. Electrodes are implanted deep in the brain, targeting the ventral portion of the anterior cingulate cortex, in Brodmann's area 25.

Other brain regions have been targeted for DBS in major depressive disorder, including the nucleus accumbens, but today we'll focus on the work of Dr. Helen Mayberg and her colleagues at Emory University in Atlanta, Georgia.

Figure 1 (Holtzheimer et al., 2012). Surgical targeting. Preoperative MRI shows the sagittal (A) and coronal (B) views of the planned optimal subcallosal cingulate (SCC) white matter target (red circle). The dotted black line indicates the subcallosal plane of interest, parallel to the anterior-posterior commissural line; the dotted white line indicates the rostral limit of the subcallosal plane; and the dotted red line indicates the midsubcallosal plane. The red circle indicates demarcation of the SCC white matter target and surrounding gray matter. C and D, Postoperative computed tomography scan merged with preoperative MRI showing a typical case with the deep brain stimulation electrodes in situ. Note that the contacts span the SCC gray and white matter in the vertical plane proximal to the split of the cingulum bundle and rostral medial frontal white matter tracts (C, red arrows, sagittal view). Contacts are numbered by convention (1-4 on the left, 5-8 on the right), inferior to superior. Contacts 2 and 3 are directly in the SCC white matter, and contacts 1 and 4 are in the inferior and superior gray matter, respectively.

Why stimulate subcallosal cingulate/area 25 1 in depression? Previous neuroimaging studies by Mayberg and colleagues (2000) showed that resting glucose metabolism in this region is overly active in depressed people, and a reduction in activity was associated with antidepressant treatment response. Another key observation was made using a mood induction paradigm in healthy volunteers (Mayberg et al., 1999). After the participants remembered a sad autobiographical memory, their SCCs showed greater blood flow relative to a neutral mood state. Thus, the "sad cingulate" was implicated in normal sadness as well as in depression.

The most recent DBS report, published in the Archives of General Psychiatry (Holtzheimer et al., 2012), is a follow-up after two years of chronic, high frequency stimulation of the subgenual cingulate white matter. The basic findings have been summarized elsewhere, including Providentia, with a review of possible mechanisms at The Scicurious Brain.

The aspect of the study that I'd like to focus on today is the inclusion of patients with Bipolar-Type II (BP-II) for the first time, in addition to those with unipolar depression. Just as with the unipolar patients, those with BP-II had to be in a depressive episode for at least 1 yr.

The specific DSM-IV diagnostic criteria for Bipolar II Disorder are:
A. Presence (or history) of one or more Major Depressive Episodes.

B. Presence (or history) of at least one Hypomanic Episode.

C. There has never been a Manic Episode or a Mixed Episode.

D. The mood symptoms in Criteria A and B are not better accounted for by Schizoaffective Disorder and are not superimposed on Schizophrenia, Schizophreniform Disorder, Delusional Disorder, or Psychotic Disorder Not Otherwise Specified.

E. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning.

Are the same neural circuits implicated in treatment-resistant depression also involved in BP-II? Remarkably, there is nothing in the literature that presents a rationale for using DBS for bipolar depression specifically, nor about why the subgenual cingulate white matter should be the target. A 2010 review by Lipsman, Lozano, and others from the Toronto neurosurgical group stated: "There are currently no trials or reports in the literature on the use of DBS for the exclusive treatment of bipolar disease or bipolar depression." However, Mayberg's ongoing clinical trial at Emory has included individuals with Bipolar Type II (current episode depressed) since it began in 2006. Those with Bipolar I have always been excluded. Why? Is it because a DBS-triggered hypomanic state is less likely in BP-II than in Bipolar I, or is it that full-blown mania might be more easily triggered by stimulation of SCC? We don't really know whether there's any empirical evidence behind these choices.

The Holtzheimer et al. (2012) paper mentions an earlier trial (Mayberg et al., 2005; Lozano et al., 2008) that included BP-II participants by accident (n=2), saying that "neither received significant benefit from SCC DBS." In those studies, conducted by the Toronto group, individuals with bipolar disorder were specifically excluded (Lozano et al., 2008), but...
One patient who initially was diagnosed with unipolar depression had more accurately, in retrospect, previously undiagnosed bipolar II disorder.
Different medications are prescribed as first-line treatments for unipolar and bipolar depression: antidepressants vs. mood stabilizers, respectively. In fact, antidepressants are most often contraindicated in bipolar depression because of the risk of a switch to hypomania/mania (Daray et al., 2010). Shouldn't we be concerned about DBS-induced hypomania? Interestingly, that was not an issue in this study:
No hypomania or mania occurred, and there was no significant change in Young Mania Rating Scale scores in any patient. None of the instances of anxiety was associated with other hypomanic symptoms.

In their review, Lipsman et al. (2010) cautiously concluded that DBS for refractory bipolar depression might be a good idea, with the subgenual cingulate as a potential target:
Given the evidence presented here, including neuroimaging, both structural and functional, as well as circuitry data and previous experience with DBS, a case can be made to carefully evaluate neurostimulation in highly selected cases of treatment-resistant bipolar depression. A convergence of data suggest that the anterior cingulate, and specifically the subgenual cingulate, should be a leading target for intervention, as it is in the best position to modulate cortico-limbic circuits mediating depressed mood. It also happens that the sgACC has been postulated as interacting in circuits responsible for impulsivity and mania, and an effect of DBS on irritability and agitation could be monitored as well (95, 96). There is sufficient unmet clinical need and a strong scientific rationale for a trial to address the utility of DBS in refractory bipolar depression and to determine whether this is a viable and safe alternative for a patient population with no other treatment alternatives.
Finally! We did find a rationale, but looks like it's a little late in the game.2

What have we learned so far? Subcallosal DBS for bipolar depression is relatively safe and does not appear to trigger hypomania. But is hypomania the opposite of sadness? Not exactly. Although an expansive and elated mood can be one feature of hypomania, it isn't a necessary one. The mood can be irritable instead. Other symptoms can include alterations in cognition (racing thoughts, distractibility, grandiosity), impulsivity, and an excessive pursuit of rewarding activities without concern for the consequences (i.e., spending the kids' college fund to buy yourself a luxury speedboat). The latter symptom might be more closely related to abnormal activity in the brain's reward circuitry. Indeed, mania-like episodes have been triggered in patients with Parkinson's disease and OCD, who have electrode placements in dopamine-rich areas. Readers who are skeptical that behavioral "addictions"/compulsions3 can exist should read the literature on dopamine agonist therapy and subthalamic DBS in Parkinson's patients.

In cases of unipolar depression, the nucleus accumbens (NAcc) has been another region of interest. It makes sense as a DBS target area from the standpoint of anhedonia (inability to experience pleasure from normally pleasurable life events) in major depression. Why not stimulate the "pleasure center" when you're feeling blue? Extensive research in animals and humans has demonstrated "hedonic hot spots" in the NAcc. However, this could be a disaster in cases of bipolar depression.

As we've seen in a number of studies, quieting the overly active "sad cingulate" can ameliorate depressive symptoms. I would be interested in seeing the details of the clinical ratings on instruments such as the HRSD {PDF}. Did the deeply sad mood and suicidal thoughts show the greatest improvement? In cases of NAcc stimulation, did symptoms of anhedonia improve the most?

Returning to the present DBS study, the clinical efficacy results are summarized in the table below. Note that the dropout rate over 2 yrs was higher for the bipolar (4 of 7) than for the unipolar participants (2 of 10). This was not adequately explained.

ADDENDUM Jan 21 2012: The first author of the article, Dr. Paul Holtzheimer, clarified that not all patients had reached the 2 yr follow-up point yet. 24 weeks of active DBS was the primary efficacy end point reached by all 17 patients. Only one participant (unipolar) had dropped out of the study.

Nonetheless, among the 11 patients who received 2 yrs of chronic stimulation, the response and remission rates were high (92% and 58%, respectively). Providing such an improved quality of life in a population that had experienced misery and poor functioning for so long is an achievement.

In the movie Melancholia, Justine is buoyed by learning the world is about to end. Now,4 an alternative ending may exist to lift people out of an apocalyptic abyss.

Further Reading:

A Depression Switch?

The Sad Cingulate

Sad Cingulate on 60 Minutes and in Rats

Good News/Bad News Update on Nucleus Accumbens DBS for Treatment-Resistant Depression

Mania and Artistic 'Surprise' Induced by Deep Brain Stimulation


1 The terms subcallosal cingulate, area 25, and subgenual cingulate have been used interchangeably in the literature.

2 Mayberg used to be part of the Toronto group but is currently at Emory.

3 Also known as impulse control disorders (e.g., for gambling, sex, or shopping). Whether these are properly known as addictions is a matter of debate. The point here is that these behaviors can be initiated through dopamine therapies.

4 In rare, selected cases.


Daray FM, Thommi SB, Ghaemi SN. (2010). The pharmacogenetics of antidepressant-induced mania: a systematic review and meta-analysis. Bipolar Disord. 12:702-6.

Holtzheimer, P., Kelley, M., Gross, R., Filkowski, M., Garlow, S., Barrocas, A., Wint, D., Craighead, M., Kozarsky, J., Chismar, R., Moreines, J., Mewes, K., Posse, P., Gutman, D., & Mayberg, H. (2012). Subcallosal Cingulate Deep Brain Stimulation for Treatment-Resistant Unipolar and Bipolar Depression. Archives of General Psychiatry DOI: 10.1001/archgenpsychiatry.2011.1456

Lipsman, N., McIntyre, R., Giacobbe, P., Torres, C., Kennedy, S., & Lozano, A. (2010). Neurosurgical treatment of bipolar depression: defining treatment resistance and identifying surgical targets. Bipolar Disorders, 12 (7), 691-701 DOI: 10.1111/j.1399-5618.2010.00868.x

Lozano AM, Mayberg HS, Giacobbe P, Hamani C, Craddock RC, Kennedy SH. (2008). Subcallosal cingulate gyrus deep brain stimulation for treatment-resistant depression. Biol Psychiatry 64:461-7.

Mayberg HS, Brannan SK, Tekell JL, Silva JA, Mahurin RK, McGinnis S, Jerabek PA. (2000). Regional metabolic effects of fluoxetine in major depression: serial changes and relationship to clinical response. Biol Psychiatry 48:830-43.

Mayberg HS, Liotti M, Brannan SK, McGinnis S, Mahurin RK, Jerabek PA, Silva JA, Tekell JL, Martin CC, Lancaster JL, Fox PT. (1999). Reciprocal limbic-cortical function and negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry 156:675-82.

Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM, Kennedy SH. (2005). Deep brain stimulation for treatment-resistant depression. Neuron 45:651-60.

A beautiful movie about the end of the world

Melancholia is also the name of the giant blue planet that destroys the Earth.

Subscribe to Post Comments [Atom]

Saturday, January 14, 2012

Remembering and Forgetting in Traumatized Ugandan Refugees

Gulu, Uganda (vis photography)

Most of us have memories from the past that we'd rather forget. When those memories are of a traumatic nature, they can more difficult to expel from our minds. Unwanted memories can be rejected by means of active inhibitory processes (Anderson & Levy, 2009), but these mechanisms are impaired in individuals with post-traumatic stress disorder, or PTSD (Zwissler et al., 2011):
Essentially, PTSD patients have trouble remembering what they are supposed to remember and forgetting what they would rather not remember. They appear to have impaired memory control.
A group of German investigators conducted a study on memory and forgetting in one of the more unsettling regions of the world: northern Uganda. The Lord's Resistance Army (LRA), a terrorist organization, has waged a long and brutal campaign to overthrow the government of Uganda:
Rape, torture, and murder have become the group's hallmarks in the almost fifteen [twenty or twenty-five] years that they have terrorized the citizens of Northern Uganda. The ranks of the LRA are filled in large part (approximately 80%) by children, who are kidnapped and brainwashed into service with the group. Human rights NGOs place the number of children currently fighting with LRA at around 3,000. LRA members also kidnap children, particularly girls, to serve as sex slaves; some have even been given as "gifts" to arms dealers in Sudan.
Zwissler and colleagues (2011) recruited severely traumatized participants for a study on directed forgetting, a memory task where instructions are given to remember some items but to forget others during the encoding phase. The participants were 51 young people (mean age=20.8 yrs, range 16–30) living in Internally Displaced Persons (IDP) camps near the city of Gulu in Northern Uganda. All were equally exposed to traumatic events such as abduction, but only 26 were diagnosed with PTSD, an anxiety disorder marked by intrusive memories and flashbacks.

The participants had two years of education on average, and many were functionally illiterate. For this reason, pictures were used as the stimuli (instead of words, which are commonly used in this type of study). The pictures were neutral in valence to examine whether memory issues in this population would extend to non-emotional material.

In the experiment, 28 pictures (Set A) were presented during an initial encoding phase. Each picture was followed by a symbol that signaled whether the preceding picture should be remembered or forgotten. During the test phase, all 28 pictures were presented, along with new pictures that served as "lures" that were similar to the initial set (Set B; see below).

Fig. 1 (Zwissler et al., 2011). Illustration of the picture sets showing three representative target-distractor pairs: (a) set A; (b) Set B. (Original photographs were shown in colour.)

For each stimulus, participants were told to indicate whether they had seen it before, regardless of the prior instruction to remember or forget. Overall accuracy in the task is shown in the figure below. The non-PTSD group had better memories for the pictures they were told to remember, compared to those they were told to forget. In contrast, the PTSD group showed no difference in accuracy for the to-be-remembered vs. the to-be-forgotten pictures.

Fig 2 (Zwissler et al., 2011). A comparison of the effect of directed forgetting on discrimination accuracy in the two groups. PTSD, post-traumatic stress disorder. *Indicates significant differences.

One way to view these results is that the participants with PTSD performed worse than controls for items they were supposed to remember, and were unable to invoke inhibitory processes to suppress memory for the to-be-forgotten items ("trouble remembering what they are supposed to remember and forgetting what they would rather not remember"). Breaking down task performance a little further, the PTSD group was more inclined to make "false alarm" errors to the lures related to pictures they were supposed to remember. This suggests that the details of the to-be-remembered pictures weren't encoded as well, and were more easily confused with related pictures they didn't see.

The authors concluded that...
...traumatized individuals with (but not without) PTSD are impaired in their ability to selectively control episodic memory encoding. This impairment may contribute to clinical features of the disorder such as intrusions and flashbacks.
However, "directed forgetting" is usually not a practical strategy when real life events are unfolding. Do these results this imply that the non-PTSD group was better able to dissociate themselves from traumatic events as they were occurring (or shortly thereafter)? Whether such a process can effectively occur at all during horrible tragedies is highly controversial (e.g., Terr vs. Loftus). The phenomenon is more often studied when applied to the retrieval of traumatic or unwanted memories (Anderson & Levy, 2009), not during the encoding phase.

Tragically, there appears to be No End to LRA Killings and Abductions in central Africa, according to Human Rights Watch. These ongoing atrocities should not be ignored.

So watch the video Dear Obama: A Message from Victims of the LRA.

Further Reading on Forgetting:

Forgetting is Key to a Healthy Mind
Letting go of memories supports a sound state of mind, a sharp intellect--and superior recall

Living and Forgetting

I Forget...

I Forgot...

...and it's Memory Week at the Guardian.


Anderson MC, Levy BJ. (2009). Suppressing unwanted memories. Curr Dir Psychol Sci. 18:184-194.

Zwissler, B., Hauswald, A., Koessler, S., Ertl, V., Pfeiffer, A., Wöhrmann, C., Winkler, N., & Kissler, J. (2011). Memory control in post-traumatic stress disorder: evidence from item method directed forgetting in civil war victims in Northern Uganda. Psychological Medicine, 1-9 DOI: 10.1017/S0033291711002273

"This was taken at the Guru Guru IDP camp. It was definitely the most uncomfortable place we visited with the most obvious results of 22 years of war and spiritual oppression. Favor of God Ministries is just beginning to work in this camp. They begin with 2 weeks of Trauma Counseling and then offer a 2-month long Portable Bible School. If it’s like any of the other camps we visited, it will be a completely different place afterwards."

vis photography
by Leah Vis

Subscribe to Post Comments [Atom]

Friday, January 06, 2012

Subjects Wanted to Drink Bourbon and Watch Erotic Films

Our fun New Year's Eve post reviewed the suspected brain mechanisms of an alcohol blackout, or an episode of amnesia after a bout of heavy drinking (Rose & Grant, 2010). Alcohol-induced alterations of hippocampal circuits are thought to disrupt memory encoding, which can lead to two different types of blackout: en bloc, a complete loss of memory for the affected time period; and fragmentary, where bits and pieces of memories remain. The en bloc blackout is more likely to occur when a large quantity of alcohol is ingested in a short period of time.

In 1970, less was known about the causes and mechanisms of alcohol blackouts. A study by Goodwin and colleagues set out to learn more:
Some psychiatrists believe blackouts to be a functional disturbance, related to guilt or anxiety. Others believe they reflect a toxic effect of alcohol on the brain. There are few data to support either concept. Blackouts usually occur erratically; comparable amounts of alcohol do not always produce memory loss, and it has not been possible to predict when an intoxicated person will suffer amnesia. During the amnesic interval, the person may function reasonably well and perform complicated acts, suggesting that the amnesia is retrograde. To confirm this, however, memory must be observed systematically during drinking periods followed by amnesia, which heretofore has not been done.
Human subjects protection programs were less stringent (or non-existent) 40 years ago, so it was no problem to enlist unemployed alcoholic day laborers to drink a pint of bourbon and undergo memory testing of an unusual sort.
The subjects were recruited from individuals seeking daily employment at the Casual Labor Division of the Missouri Employment Office in St. Louis. Previous experience with this source had shown that a sizable proportion of the job applicants are alcoholic. Individuals were eligible for the study if they met two criteria; (a) willingness and ability to drink more than a pint of whiskey in a few hours, and (b) good physical health.

Ten men (mean age 41 yrs) were recruited as participants. Eight of them were alcoholics, and five experienced frequent blackouts. After fasting for 5 hrs, the experimental procedure was to drink 16-18 ounces of 86 proof bourbon on the rocks in 4 hrs. Cognitive testing began after the first hour of drinking to assess immediate, short-term, and remote memory as well as calculation abilities. In brief, the drunk participants looked at toys and watched porn and were asked to recall these items after a delay (Goodwin et al., 1970):
The tests were as follows. (a) Every 30 min the subject was shown a toy for 1 min, after which it was removed from sight. Two minutes later the subject was asked to recall the toy. If recalled correctly immediate memory was considered intact. Thirty minutes later he as asked again to recall the toy to measure short term memory...

(b) Every 30 min the subject was shown for 1 min a scene from an erotic movie. He was asked to describe and discuss the scene. If he could recall the scene 2 min later, this was considered a reflexion of intact immediate memory. Thirty minutes later he was again asked to recall the scene to measure short term memory.

(c) Remote memory was tested by asking questions about early upbringing (name of schools, teachers, and so on) and events that occurred during the previous 2 days.

(d) Ability to perform calculations was tested by asking the subject to do simple multiplication and subtraction tasks.

Other than wanting to create a highly desirable means of employment for the target population, why did the authors use erotic movies as stimuli?? (1) So that memory for arousing, emotional material could be compared to memory for the [presumably] more neutral toys; (2) Because very drunk men would be unwilling or unable to cooperate if more boring tests were used; and (3) Toys and porn were highly memorable to sober male subjects 24 hrs later.

The results indicated that all 10 subjects performed well when tested at the 2 min retention interval, but half of them showed impaired memory when tested 30 min and 24 hrs later. These same 5 participants were the ones with histories of blackouts. The authors speculated that the faster rise in blood alcohol levels (BAL) in the blackout subjects could be a reason for their amnesia. However, this explanation is implausible because the blackout group had poor memories at the earliest 30 min time point, when there was no difference in BAL between the groups.

So what have we learned? Retrieval of remote memories and simple calculation abilities are not impaired by heavy drinking, and... is possible to predict amnesia during a drinking bout: on formal testing individuals who are later to experience amnesia will have a specific loss of short term memory correlated in time with the period of amnesia, while other types of memory are intact.
A final lesson is that it was much easier to publish in Nature in 1970. Group comparisons with n=5? Fine and dandy. Pesky statistics and error bars on figures? Who needs them? Those were the days my friend...


GOODWIN, D., OTHMER, E., HALIKAS, J., & FREEMON, F. (1970). Loss of Short Term Memory as a Predictor of the Alcoholic “Blackout” Nature, 227 (5254), 201-202. DOI: 10.1038/227201a0

Rose, M. & Grant, J. (2010). Alcohol-Induced Blackout. Journal of Addiction Medicine, 4 (2), 61-73.

Subscribe to Post Comments [Atom]

Monday, January 02, 2012

All about the brain and its workings.

Thanks to Tyler Cowen, Professor of Economics at George Mason University and an author of the blog Marginal Revolution, for his Very Short Review of this blog in the New York Times Magazine.

Subscribe to Post Comments [Atom]

eXTReMe Tracker