Rationale for Csillag-Reiner Terminology (6/02)

We offer the present commentary as a rationale for the Csillag-Reiner terminology, which is presented as two attached Microsoft Word Tables. One table deals with Basal Ganglia and related cell group terminology and the other with pallial terminology. They are formatted in letter horizontal page orientation at 75%, if you need to format them upon opening. In the course of stating our principles, our comments will unavoidably reflect on the principles guiding the other terminologies offered. There are, in fact, however, many commonalities in the principles underpinning specific proposals and many similarities in some of the actual terms. There are also some differences. We believe that detailed discussion of the merits of specific proposals and the principles guiding them will be the province of the nomenclature conference. This commentary and accompanying tables are offered as a set of ideas and suggestions to consider.

General Comments

The purpose of the terminology change is to achieve a nomenclature that is consistent with current knowledge of bird brain organization, and facilitates communication among neuroscientists, regardless of whether they specialize in avian brain or not. We as a group are currently focusing our efforts on the telencephalon and some allied brainstem cell groups. As we all know, many of the terms traditionally applied for telencephalon and some related brainstem areas imply incorrect homologies to mammals. In some cases, the homology is wrong at the one-to-one level, in other cases at the regional level. This hinders the impact of our field and the dissemination of our findings.

We believe that for the sake of continuity and ease of transition, it is important to devise a new terminology that can be readily learned and easily linked to the old terminology. For much of the basal ganglia system, we believe that switching to established mammalian terms is warranted, will provide us with new names that can be easily remembered, and should be easily linkable to the older literature since the number of terms to change is not vast. Some of the proposed new terms are, in fact, already being used in published studies. This proposed change (to some extent a formalization of what some, such as Tony Reiner, are already doing in their published work) would make the avian literature easier to read for an outsider, and thereby aid the impact of findings on avian basal ganglia to understanding of basal ganglia function (or connectivity or development) in general.

For the pallial terminology, the avi-community will have a choice among four basic options. One option is to keep what we have. We would only embrace this if we did not like any of the other three options. A second choice would be to use this opportunity to completely revise the pallial terminology (fiber bundles and lamina included) using a rational, descriptive terminology. The Medina-Paxinos-Puelles terminology is an example of this approach. The disadvantage of this approach is that a drastic change would require researchers to be conversant in two terminologies to be able to assimilate the old and new literatures. A third approach would be to develop new terms similar to the old terms (but anatomically correct and informative) that retain the same abbreviations. This is the approach of the Csillag-Reiner terminology. The difficulty with this approach lies in devising credible new terms within the constraints of the old abbreviations. A final approach involves eschewing names and simply using the existing abbreviations for the various problematic pallial structures. Ann Butler first brought up this approach, and Georg Striedter and David Perkel presented a pallial terminology of this type. The problem with this approach is that it is essentially a short-term compromise position, since a letter-based terminology is likely to prove difficult to learn for newcomers and outsiders.

An additional issue to think about is if we want a classical language-based terminology that can be readily Anglicized, or if we want one that is Anglicized from the start and as a consequence might not be easily reversed to a classical terminology.

For the pallium, where we believe only general homology at the level of a few fundamental pallial subdivisions is currently possible, we recommend retaining existing abbreviations as much as possible. Because many of the detailed homologies are uncertain, we cannot change to mammalian terms (which might be inappropriate in any case given the divergent evolution of the avian pallium). If we change extensively to a novel descriptive terminology requiring new abbreviations, it would be problematic for newcomers to the field to have to learn two sets of terms and abbreviations - one to read the old literature and one to read the new literature. It would be burdensome for established researchers as well. Moreover, it would make avian neuroanatomy yet more daunting to outsiders to the field, such as those studying mammalian brain.

Devising new terms for the pallium that retain old abbreviations, by contrast, would avoid these problems and also result in automatic new names with the same old initials for many specialized structures that are built upon the Karten-Hodos terminology - such as LMAN and RA in songbirds. Having abbreviations linked to actual names will make them easier to learn and recall than would a pure letter nomenclature.

Basal Ganglia Terminology

Our table presenting our Basal Ganglia terminology includes a few minor alternatives for spellings and names. There seems little basic disagreement among the various proposals about the specifics of a terminology change for the basal ganglia and its related cell groups. We will merely restate, for the sake of thoroughness, the points about basal ganglia terminology previously made in Tony's email. For those who have already read them, the next few paragraphs are only slightly edited versions and can be skipped.

Erich has raised the concern that LPO and PA are not strictly medial to lateral in orientation, and so a simple name change to medial and lateral striatum, respectively, might be inappropriate. Erich notes that LPO is actually rostromedial to PA and so names recognizing this might be more appropriate. In this regard, we would caution against developing terms that are too complex and therefore not "extensible" (invoking the term and notion stated by George Paxinos at the avi-dinner at SFN2001). All of LPO is not rostromedial to PA, and so this terminological approach is not really much more accurate than calling them medial and lateral striatum. We would let simplicity rule here, as do Luis, Loreta, George, Georg and David.

Note that (in contrast to Georg and David but like Loreta-George-Luis), we believe the PP needs to be called the dorsal pallidum, not just the pallidum, to distinguish it from the limbic pallidal territory typically called the ventral pallidum (often called ventral paleostriatum in the avian literature of the 1970-1990s).

Georg and David generally recommend avoiding renaming diencephalic structures. We agree, except that we believe the evidence is overwhelming that ALa is the subthalamic nucleus (as do Loreta-George-Luis) and such a renaming would be useful. If the avi-eaters regard this change as premature, then perhaps we can agree to use "avian subthalamic nucleus" as a somewhat informal alternative to ALa.

While we agree with the various comments and suggestions about nucleus accumbens in birds being some rostromedial part of LPO, we do not actually know the boundaries of this region, and so its borders cannot be specified in any atlases as of now.

Loreta-George-Luis recommend that the medial and ventrocaudal part of LPO be called the medial pallidum. We believe that this is problematic for several reasons. First, the evidence that medial ventrocaudal LPO is pallidal in adult birds is weak. The notion that ventrocaudal LPO is pallidal is based on hypotheses about the mediolateral developmental extent and the extent of adult retention of the pallidal sector of the subpallium, and on an interpretation of a cadherin discontinuity in Redies et al. (JCN2001). Thus, the locus, extent, and very existence of this pallidal LPO sector in adult birds are not truly known. Secondly, Loreta-George-Luis acknowledge that medial ventrocaudal LPO has striatal traits (which it does), but suggest it has pallidal traits as well due to its presumed developmental history as a pallidal sector. Because of the admitted presence of striatal traits (in fact, striatal traits predominate), it would be misleading to call medial ventrocaudal LPO pallidal. It seems safer to us to include medial ventrocaudal LPO as medial striatum for now. If at some future time we have a clearer understanding of the possible admixture of striatal and pallidal neurons in some part of LPO (perhaps as a forerunner of area X) and the developmental history of this region, then this can be recognized by redrawing some boundaries and deciding on a name suitable to what we have by-then learned of the adult structures to which we are applying names.

Pallial Terminology

For our pallial terminology, we recommend names that allow the retention of most of the existing Karten & Hodos abbreviations. The chief challenge is devising new terms that are euphonious, descriptive and accurate. The avi-eaters will need to be the judge of our success. In our table, we present our basic abbreviation-preserving suggestions, as well as various alternatives. Below we comment on our suggestions, and reflect on the merits of the various alternative choices.

We will repeat a few points about archistriatum from Tony's prior email message. First, as outlined by Erich at the avi-dinner at SFN2001, what has been termed Aa, Aid, Aiv, Am and Ap in the work of Zeier and Karten is clearly all pallial and none of it is subpallial by several lines of evidence. First, the work of Luis and his colleagues show this territory expresses pallial markers during development. Secondly, the region that Luis and coworkers have called Aa and said to be subpallial is clearly not the same region that Karten and Zeier or Karten and Hodos called Aa. Rather, Luis' Aa, as clearly defined in the recent Redies et al. (2001) paper in JCN, resides in the ventrolateral part of the basal ganglia (as noted by Erich at the SFN2001 avi-dinner). Thirdly, Erich's analysis of glutamate receptor localization and neurotrophic factor localization (as presented at SFN2001) shows the archistriatum to be pallial. Finally, Tony's own recent work using multivariate statistics to analyze diverse neurochemical traits of the avian archistriatum and mammalian amygdala show archistriatum (i.e. Karten's Aa, Aid, Aiv, Am and Ap) to be pallial in nature. Tony will be presenting this work at SFN2002, and can also review it at the nomenclature conference, if this is deemed appropriate or necessary. Thus, in contrast to the concerns expressed by Georg and David in their commentary on their terminology, we believe it valid to replace -striatum with -pallium in changing the name of the structures that have been considered part of the archistriatum. There is, however, no need to rename the taenia at this time. We do agree it is amygdaloid in nature (Greg Ball summarized the evidence well in his email), but as noted by him its precise mammalian homologue is yet uncertain. We believe it best to await clarification of this issue before renaming the taenia.

In a commentary on whether the archistriatum possess a subpallial co

mponent, Luis presented the view that he believed it is likely that birds possess a correspondent of the subpallial amygdala of mammals, and this view partly guides his quest for the location of such a region. We entirely agree with the premise that such a region may exist and is important to look for, but we note again that what has been called a subpallial part of the archistriatum in Redies et al. (2001) resides below the LMD and does not reside in any part of what Karten and Hodos called the archistriatum.

Should we call the archistriatum the amygdala, as in the Medina-Paxinos-Puelles terminology? While we are largely in agreement with Loreta-George-Luis that archistriatum is likely to be homologous in some way shape or form (which may preclude one-to-one equivalences) to mammal pallial amygdala, we (like Georg and David) do not think that we are yet so sure that we should call avian archistriatum the avian amygdala. Moreover, some parts of archistriatum (i.e. Ai) have diverged quite far from amygdaloid connectivity (assuming they were ever amygdaloid), since they are somatic rather than visceral in their connectivity. Calling them amygdala might be misleading for this reason as well.

Archistriatum we recommend be called the arcipallium in light of its arched contours and its pallial nature. We have been toying with other prefixes or words that begin with "A" that could serve as alternative possibilities. In light of the uncertain nature of the archistriatum it could be called nucleus ambivalens, or the ambipallium. (since it seems both somatic and visceral). Or since it is a different type of pallium, it could be called the allopallium (meaning other pallium). The term allopallium, however, might convey some unwanted association with allocortex in mammals (i.e. hippocampus and pyriform cortex). We'll keep thinking about credible "A" alternatives, and welcome suggestions.

For hyperstriatum, we recommend hypsopallium, with hypso- being a Greek prefix that means upper. Hypso- appears in the name of a few animal species, so it has some currency in biology. The first option we present in our table, then, is to replace hyperstriatum with hypsopallium. We recognize, however, that the terminology change effort has provided an opportunity to dissociate the HV from the parts of the Wulst with hyperstriatum in their name by assigning the Wulst structures and the HV different root names. We see that many favor this option. The second choice we give in our table involves this approach. Renaming HA, HIS and HD as parts of Wulst is not too radical a departure from the status quo; and so it might be a suitable way to rename these structures without becoming disengaged from the past literature. The key question to us about this is the positional adjective to be used along with the root term Wulst. The radial developmental unit idea driving the Medina-Paxinos-Puelles renaming of HA-HIS-HD along a medial-to-lateral axis leads, we believe, to a problematic terminology. We do not believe that developmental topology/topography should invest an adult nomenclature when this approach causes structures that are not medial (e.g. HA in owl) to be called medial, and structures that are not lateral (e.g. HD in owl) to be called lateral. We are after an adult terminology that describes position in adults, irrespective of species differences or transient developmental topographies