Sunday, September 5, 2010

Bedside testing of frontal lobes

Kramer JH, Quitania L in Miller BL, Cummings JL.  The Human Frontal Lobes functions and disorders.

Published tests:
EXIT-  Executive Interview

bedside tests
fluency -- give F, A, S test and in alternate form give D only. Measure not only number of words produced, but also rule violations (giving responses with a different letter, and proper nouns) and repetitions.  Subjects who do worse on category fluency have more likelihood of temporal involvement, and those with letter fluency only are more likely to have frontal involvement.  Nonverbal analogue to letter fluency is design fluency. Show a box with five dots and ask them to create in one minute that have four lines that connect the dots. Several variations are used, one is d-kefs.

Mental flexibility-- on Trails B, Stuss et al. enhance the test by counting errors on Part B, which correlates best with DLPFC right sided injury. On DKEFS Trails, subjects alternate with letters and numbers.  Older patients maybe asked to shift between numbers and days of the week, with a record of time to complete task and how many sequencing errors are made.  Patients who cannot read may get Color Trails instead, connecting colored circles alternating between pink and yellow circles.  Alternate form of Trails:  Patients get one minute to create, from an array of dots (alternating half filled dots, half unfilled), as many designs as possible with four straight lines connecting dots.  In first form, they are asked only to use unfilled dots, in the second, to alternate between filled and unfilled dots. Design and rule violations (ie failure to shift) are recorded.

Abstract reasoning:    a) similarities between word pairs (dog/lion, table/chair. anger / joy)   and b) proverbs- eg. " An old ox plows a straight row" and "Shallow brooks are noise" and " A beard well lathered is half shaved."  Give credit for abstract and accurate answers.

Response inhibition:  a) Stroop-- tabulate correct responses in 60 seconds, and the number of inhibition errors.  b)  Salient responses :  Tell patient to to point to his chin while examiner touches nose, or tell patient to point to the ceiling while extending his hand   c) Opposite-- when I tap once, you tap twice, and vice versa.

Working memory-- backwards digit span, spelling "world" backwards, serial sevens, saying days of week backwards, months in reverse order.

primitive reflexes-- snout, suck and grasp and glabellar

Complex motor programs
1) fist palm edge task-- perform 10 times, record the number of trials to learn the task, the number of perseverations, and the fluency of the movements.
2)  Graphomotor sequences-- alternate m and n (cursive) elicit perseveration, micrographia.
3)  Rhythm test-- ask patient to imitate finger taps in complex patterns.

Behavior-- observe and record, use a modified Manchester Behavior Questionnaire to survery based on both information and report.  NPI and NPI-Q (short version) can be used (link to certification test: use the Dysexecutive Questionnaire (DEX), Frontal Behavior Inventory (FBI) , and Frontal Systems Behavior Scale (FrSBE)

Saturday, September 4, 2010

New Executive Brain VII Summary

Believes hippocampi is "center" for veridical decision making, DLPFC for adaptive decision making. 

Ventral tegmental lesions causing pseudofrontal syndrome

Goldberg et al. Cortex 1989  Reticulofrontal disconnection
update see Miller Nat Rev Neurosci 2000

DA depletion in midbrain caused frontal disconnection.
"Kevin" had retrograde without anterograde amnesia

Analogy- (partial) to schizophhrenia that also affects mesolimbic (temporal) and mesocortical (frontal) projections.  EG argues schizophrenia represents a top down organizing defect ie. a type of auditory agnosia.  Behav Brain Sci 1991.

New Executive Brain VI: Syndromes Orbitofrontal (OF) Pseudopsychopathy

Term is discarded.  However, OF lesions inject emotion into analytic decision making, prevent responding like an automaton. 

Social interactions:  ACC reins in amygdala.  In monkeys, ACC is more important than OF in socially appropriate behaviors (Ito et al.). 

EG argues, about moral reasoning and the frontal lobes, that understanding temporal relationships is key to understanding causation, and appreciation of consequences of behavior is necessary for moral reasoning.  Inferential reasoning of the if.then kind, is at basis both of language and of moral reasoning.  Counterfactuals, that is, if I had done this x then y would have happened, are impaired after OF damage.

New Executive Brain 3: Frontal lobe tests

Real life memory is not memorizing list given to us, but selecting task, and learning contextual information to task we chose.  Therefore the memory task is not "working memory" any longer. This may explain the disparity in frontal lobe role in memory in humans and animals.  Klingberg "The Overflowing Brain" (2008) gives a nice rendition of state of working memory.

The New Executive Brain V? Emotional brain, basal ganglia etc

Davidson et al. Left hemisphere lesions produce negative affect especially if they are closer to the frontal pole.  Right hemisphere lesions produce euphoria.  Left hemisphere activates to happy film, financial gain, right hemisphere to sad film and financial losses.  In the Ultimate Game, the right hemisphere activated to unfair offers.  Suppression of memories occurs via right or bilateral DLPFC affecting the right hippocampus.Meditation, deactivates left frontal.  Vicarious pain (watching others suffer) activates right frontal.  (Naomi Eisenberger,et al. Science 2003, study of social exclusion). 

Frontal lobes modify attenuate the amygdala (the "pre" prefrontal lobe actor); this done through orbitofrontal cortex.  The amygdala also show evidence of valence mirroring that of evolutionarily later frontal lobe, ie left positive valence, right negative.  Suppressing negative emotions is left DLPFC on left amygdala, fear engendering stimuli affect right amygdala, and lesion of right amygdala reduces the ability to experience fear.  Anxiety may be associated with a large or hyperactive right amygdala.  Emotionally charged crimes may be related to large right amygdala.  Removal of right amygdala may attenuate ability to appreciate emotion of fear in others' faces. 

In gambling experiment, both prefrontal and amygdalaoid lesions interfered with decision making, but only amygdaloid lesion was associated with a change in skin response.  Distant threats in a virtual reality environment were mediated by DFPFC, but immediate threats were mediated by amygdala. 

Author argues novelty/familiarization dichotomy can tie up loose ends of verbal/nonverbal debate. 

Parkinson patients avoid choices leading to bad outcomes (risk aversion) more than selecting ones leading to a good outcome.  (Frank et al., Science 2004). 

The New Executive Brain 3: Tests of Frontal Function

In real world, we choose a path and learn information we need; in memory tests we are given a list of words and told to memorize.  Difference is role of frontal lobes.  Hence, animal experiments show a more pronounced role for frontal memory than human experiments.  CVLT allows semantic clustering, which (postulated) may be due to frontal activity.  Most working memory tests lack a selection component.   Frontal role is to select information and hold it online and manipulate it as long as is needed.

Courtney et al. (Cereb Cortex 1996) showed faces on a grid then another set of faces with "what" question and "where" question, which respectively, activated ventral and dorsal visual stream representations within the frontal lobes (inferior and superior). 

"How does an organism decide what information is important enough to be kept online" is a question that has been ignored in almost all neuropsychological paradigms.  Goldberg divides decision making into veridical and ambiguous.  Veridical is "I spend 300 of 1000 dollars, how much is left?"  Ambiguous or adaptive is "Shall I wear gray, blue, or white?"  What is the answer?  Gertrude Stein asked "What is the question?"  Disambiguation may requires prioritizing (remember Buridano's donkey from Luria's books).  Individual must be able to disambiguate in multiple ways and switch set ( "White is for spring, but white is not allowed at school so I will wear gray").  Veridical decisions find the truth, adaptive ones find what is best for the organism. 

In Cognitive Bias Test, individual is given a card and asked to pick the one "they like the most."  Then they were asked to find the choice "most similar to the target" then the one "most different from the target."  Frontal lobe is required only for the first condition, the only one requiring disambiguation.  Frontal lesions did not affect the second and third conditions.  In Alz disease, the disambiguating test declined much faster.  Verdejo found same in individuals with substance abuse (altered adaptive but not veridical decision making).

Males made choices on CBT in a context dependent way, that is when the targets changed, so did their choices.  Females made context independent choices, that is their choices were stable regardless of target.  This is first account of gender differences in adaptive decision making.  Males with damaged right frontal lobe performed in an extremely context dependent manner, and males with damaged left frontal lobes performed in an extremely context independent manner. In females, damage to either frontal lobe caused context dependent choices.  The Wisconsin Card Sorting Test failed to differentiate laterality of the lesion. 

The effects of posterior lesions were less robust than those of frontal lesions, but were nonetheless dimorphic.  In males, posterior temporal lesions followed frontal lesions:  left sided lesions made males more context independent, and right sided lesions made males more context dependent.  In females, posterior lesions were opposite frontal lesions: on either side, they made decision making more context independent. 
Goldberg argues that in male laterality is more articulated, but in the female, front-back differences are more articulated. 

In functional imaging experiments, when asked to process verbal information, males coactivated front and back in same hemisphere, whereas females coactivated homologous regions across hemispheres.  Males have greater functional differentiation of the hemispheres, whereas females have greater functional integration of hemispheres-- note they also have bigger callosa.  However, males have larger longitudinal fasciculi within hemispheres (see book for citations). 

An overwhelming number favored similarity over novel among humans on the CBT (unlike other primates who picked the novel object).  However, that is only true among righthanders.  Among lefthanders, many picked the choice that differed from the target, not the one similar to it.  J Cogn Neurosci 1994. EG suspects that the 9:1 balance in favor of right handers represents evolutionary needs for stability versus innovation.  Functional roles of left handers may be reversed on CBT (Ibid.). 

The New Executive Brain 2: Novelty, subroutines, and hemispheres

1. Yakovlevian torque (see Lemay, Ann Ny Acad Sci 1976) shows the right frontal lobe is wider than and protrudes over the left, and the left occipital lobe is wider than and protrudes over the right.  This present in great apes to some degree, and there are gender differences.  Other assymetries include the planum temporale is larger in left side in human and this is also true in several other primates.  Spindle cells with long axons are more numerous in the right hemisphere.  White matter is more organized in the right than left frontal lobes in children and adults (Klingberg et al., 1999, Neuroreport).  Dopamine is more prevalent in the left hemisphere, norepinephrine on the right, and estrogen receptors are more prevalent on right (rats) Sandu et al, Exp Neurol 1985.  Goldberg argues that hemispheric assymetries exist in animal species and these cannot be based on language since animals lack language.  Is language a "special case" of some larger difference?

2.  Other right - left approaches (analytic/holistic, sequential/simultaneous processing) are difficult to operationalize and are rejected.

3. Bates E argued that in children, left hemisphere brain damage is inconsequential, but in right hemisphere patients it is devastating (in Bronan SH ed, Changing Nervous System:  neurobehavioral consequences of early brain disorders, Oxford University Press, 1999).

4. History of civilisation is right to left shift; translating Vygotsky cultural historical psychology in neuroanatomical terms would reach similar conclusion

5.  In neuropsych tests, averaging across six categories of WCST, the five trials of CVLT, and 30 trials of Benton Lines is not accurate anatomically.  Should rescore old tests with that in mind. 

6.  Agnosias:  leitmotif is left hemisphere is top down process, unable to recognize unique exemplars of generic categories "associative agnosias."  Right hemisphere lesion cannot recognize exemplars as their own selves or as different exemplars of same category ("apperceptive agnosias").  see Goldberg JCEN 1990.

7.  Animals may have paw prefence but its closer to 50-50 unlike humans. 

8.  Frontal function:  divide into perseverative (difficulty with cognitive flexibility) unable to transition from one activity to another; and field dependent behavior (echopraxia/echolalia).  Traditionally these are known as 2 aspects of executive control, ability to guide behavior by internal representations and ability to shift cognitive sets with exigencies.   These may be referred to as "stability" and "plasticity" as well.  Author postulates that D1 receptors maintain low level maintenance of current cognitive regime, and D2 receptors respond only to high levels of dopamine and cause destabilizing rapid updating of the cognitive regime, favoring the right hemisphere.  On Executive Control Battery,  perseveration was twice as common following left hemisphere injury.  Echopraxia was twice as severe after right hemisphere injury.  However, these differences were present only in men.

4. Authors Goldberg and Costa postulated the right hemisphere is important to cognitive novelty and the left to cognitive subroutines (Brain Lang 1981: 144-173).  Information is processed from novel to familiar.  .  The left hemisphere information is stored in better articulated neural networks and easier to access. The right hemisphere is modified at a slower rate than the let hemisphere.  It deals with novelty better due to it contains averaged default representation containing shared  poorly differentiated features of many prior situations.  The evidence that supports include Bever and Chiarello (Science 1974) who showed untrained musicians process with their right hemispheres, trained ones with their left hemispheres.  He discusses obscure / familiar faces as being processed with right and left, (Henson et al, Science 2000).   Alex Martin et al. (Hippocampus, 1997) used PET to show that with presentation of meaningful words, nonsense words, real objects and nonsense objects, the right mesiotemporal lobes were activated with first, novel presentation, decreasing with the second exposure. The left hemisphere activation was constant.  Learning a complex motor skill activated first the right MFG, then late in training left posterior parietal cortex, left dorsal premotor cortex,and right anterior cerebellar cortex (Shadmehr and Holcomb, Science 1997).

More studies:  learning grammar A and B:  initial activation occurred in right ventral striatum, left premotor and anterior cingulum, replaced gradually by activation of right DLF and right posterior parietal (Berns et al.  1997 Science).  Raichle (cereb Cortex 1994) found on  a task of finding appropriate verbs for nouns, in naive condition activated ACC, left prefrontal, left temporal, and right cerebellum, but after practice only left medial occipital was activated.  A task with longer processing has right hemisphere to a greater degree.

5.  Constructs that are analagous to the novelty/familiarity distinction:  Exploration/exploitation ( eg. gambling task, novel learning of task and exploiting task to make money) involved right frontopolar regions and left DLF regions respectively (Daw et al. Nature 2006 and Heekeren et al. Nature 2004).  Another distinction is critic and actor; critic learns to predict future awards and actor uses the rewards to guide behavior.  The "critic"  in an fMRI study is the bilateral putamen and right ventral striatum, and the actor is the left anterior caudate.  This resonates with an older idea that the OF lobes appraise value to person and prefrontal cortex navigates the external world. 

6. Pathologic augmentation of orientating reaction (PTSD) has ep activity mostly in right hemisphere (Metzger, 2005 in Vasterling and Brewin). 

The New Executive Brain by Elkhonon Goldberg I

Goldberg reviews much more information in this book than can be summarized here, but I would like to go over certain information that would be critical to Goldberg's overall "brain view."  I also include random references from the book that are less pertinent but interesting background nonetheless.

Goldberg is a protege of Alexander Luria's, but emigrated from the USSR in the early 1970s, establishing a career in New York, academic, eventually private practice.  He opens with a reference to The Creation of Adam, a painting in the Cistine Chapel that Goldberg presents as representing an allegory of the frontal lobes. 

1.  The left inferior prefrontal cortex becomes active when presented with statements with semantic violations (:"trains are sour").  Hagoort et al., Science 2004.  fMRI study.  Language and executive functions arose together.  Language builds models, executive functions manipulates them and conducts operations on them.

2.  Goal formation is action centered about "I need" not "It is."  Therefore, formation of goals relates to the emergence of the mental representation of the self, in turn related to the development of the frontal lobes.

3.  The dorsomedial thalamus nucelus is the highest point of integration in the thalamus and has many connections with the frontal lobes.  It also maintains connections with the hippocampus (memory), the cingulum (emotion and uncertainty), amygdala (relations among individuals in species) and hypothalamus (in charge of homeostatic functions). 

4.  Goldberg speaks of the "fallacies" of modularity (the high tech revival of 18th century phrenology").  For example, the planum temporale is critical for phonologic processing, but is not narrowly dedicated.  Rather, it is critical for auditory analysis in a categorical sense (is the sound a dog barking, a cat meowing , etc.).  It activates both when phonological discriminations are made and when environmental sounds are matched with their sources.  Fusiform gyrus is associated with facial recognition.  Goldberg differentiates between a prior, preordained modularity and a posteriori, resultant modularity.  A priori means the area is hard wired for the task.  A gradiential view of neocortical organization means that certain tightly  integrated neuronal groupings  emerge over time, due to cognitive history of individual.  This means is has less uniformity, more individual differences and will change over time.  The left hemisphere, which has more dedication to "subroutines" has more modularity and the right hemisphere has less.  Goldberg published his gradiential theory of modularity in 1989 in JCEN and republished as a book chapter in 1990 (Contemporary Neuropsychology and the Legacy of Luria, 1990).  "Modularity is best applied to an evolutionary old structure, the thalamus."  Goldberg discusses the "dynamic topology" of cortex and the evolving role of the thalamus in its new inferior position in the hierarchy, which is a reason its function has been elusive.

5.  Jordan Grafman et al. showed in an fMRI study that when a cognitive task is characterized by a hierarchy of goals, the regions around the frontal lobes are activated when subjects have to keep in mind the main goal while performing subordinate goals (Nature, 1999).  This was supported by D'espositio et al.who showed that increasingly abstract goals activate increasingly anterior portions of the frontal lobes.  Anterior prefrontal cortex  has a high number of dendritic spines per cell and high spine density, making it ideal for a variety of inputs.

6.  E Koechlin et al (Science 2003) introduced the notion of a "hierarchical cascade of executive processes."   The control operates in Brodman areas 6 (premotor), 44/45 (inferior frontal) and 46 (middle prefrontal) at level of stimuli, perceptual context, and temporal episode, respectively.  Goldberg analogizes to perseveration  of elements , features and activities, respectively.