Wired for Sound: A Q&A Concerning Dolphin Brains

I took preliminary exams for my Ph.D. back in 1997. One of my committee members was Sam H. Ridgway Senior Scientist at the US Navy Marine Mammal Program. About half of the questions he asked had to do with neural organization in the bottlenose dolphin and modelling it. “DarkSyde” asked in comments about such things, so I thought I would post my answers to those pre-lim questions.

1. Discuss afferent and efferent routes between the ear and brain.

For my particular interests, the goal would be to determine a “wiring diagram” of the neural structure, including feedback loops and other possible control structures. This does not appear to have been done yet for any mammal. From the descriptions that I have read, either the overall coherence of path of activation has not been a major issue, or I am missing some important underlying background that has been assumed to be known by any reader. I tend to favor personal ignorance as a working hypothesis. I will give my understanding of my reading thus far, with the caveat that in several circumstances I suspect that some intermediary neurons or structures have been elided from the description.

Major sources:

[1] Iurato, S & MD Bari. 1974. Efferent innervation of the cochlea. In: WD Keidel & WD Neff (eds.), Handbook of Sensory Physiology (Vol. 1). Springer-Verlag, pp. 261-282.

[2] Harrison, JM & EM Howe. 1974. Anatomy of the afferent auditory nervous system of mammals. In: WD Keidel & WD Neff (eds.), Handbook of Sensory Physiology (Vol. 1). Springer-Verlag, pp. 283- 336.


Neural activity begins with the transduction of acoustic waves into electrochemical communication in the hair cells of the basilar membrane. Synaptic connections stimulate activity in sensory neurons. Apical, intermediate, and basal neurons of the cochlea innervate the cochlear nucleus. The cochlear nucleus gives rise to a variety of connections. Ipsilaterally, these connections are to the medial superior olive, the lateral superior olive (absent or reduced in dolphins and bats), the lateral nucleus of the trapezoid body, lateral pre-olivary nucleus, and dorsal and ventral nucleus of lateral lemniscus. Contralaterally, these connections of the cochlear nucleus are to the medial trapezoidal body, medial superior olive, medial pre-olivary nucleus, and dorsal and ventral nuclei of the lateral lemniscus. The medial trapezoidal body innervates the lateral superior olive of the same side. [2, pp. 284-294]

Afferent innervation of the retro-olivary group is laminar, and may indicate a tonotopic arrangement [1, p. 261].

Further connections of the cochlear nucleus include contralateral connections to the inferior colliculus. The medial superior olive makes only an ipsilateral connection to the inferior colliculus. In organisms with lateral superior olive, both ipsilateral and contralateral connections to the inferior colliculus exist. The inferior colliculus connects to both ipsilateral and contralateral medial geniculate bodies.

The medial geniculate body makes ipsilateral connection to the auditory cortex.

(Here is a good diagram of these connections in the human brain.)


The “olivo-cochlear bundle” of the cochlear nerve is efferent and has been described in man, cat, rat, opossum, pigeon, and caiman [1, p. 261]. Given the broad range of taxa, the assumption that it also exists in cetaceans seems fairly safe.

These connections originate both ipsilaterally and contralaterally. The descriptions vary with the subject animal and the researcher publishing the description.

The contralateral connections stem from an area near the accessory olive, the retro-olivary group of neurons and trapezoidal body nucleus. The ipsilateral connections stem either from the S-shaped olivary segment or the pre-olivary nuclei for one bundle, and the reticular formation of the pons and medulla for the other. The reticulo-cochlear bundle is described from rodents. [1, pp. 261-263]

The two separate sources of fibers form a single bundle near the point of exit from the brain. In the inner ear, the bundle separates into various fasciculi, joining the cochlear nerve and also the intraganglionic spiral bundle [1, pp. 263-264].

In man, acetylcholinesterase containing fibers are found in the cochlear nerve trunk, suggesting yet more links to efferent fibers [1, p. 264].

2. How have efferent fibers been identified?

The histological tool of choice seems to be the use of injected horseradish peroxidase. Ketten has discussed HRP and another technique suitable when a subject will be sacrificed, but indicates that she mainly has to rely upon a technique applicable to fixed specimens, which only allows the elucidation of one fiber per specimen, and still takes six months or so in preparation time. According to Ketten, there is no mammalian auditory system which has had the efferent pathways completely mapped out. Acetylcholinesterase is selectively stained or tagged in order to identify efferent fibers.

Degenerative studies compare normal histological specimens with those where lesions or surgery destroys certain areas of the neural system. Loss of normally seen connections at the periphery indicates that the missing fibers were efferent.

A physiological study stimulated the crossed OCB at the floor of the fourth ventricle, and found activity only at the pre-olivary nuclei [1].

3. How could efferent input be considered in designing a neural network model of dolphin hearing?

Until more neuroanatomical work is done on dolphin auditory efferent pathways, some cues will still have to be taken from work done in other species. This is a hazardous step, since neuroanatomical work in bats shows large variations in efferent wiring across genera (Bishop & Henson 1988). Some speculation is unavoidable, but this is precisely where model generation should be able to contribute to biological research, by identifying the likely areas of interest and suggesting mechanisms of testing model predictions.

In the Bishop & Henson study mentioned above, some species of bat have systematic multiple efferent synapses on the outer hair cells.

Physiological work in some species has not shown any change in absolute thresholds when the efferent fibers to the auditory system are cut. However, there does appear to be a loss in frequency discrimination capability and also in raising the required signal-to-noise ratio for threshold of hearing (Iurato & Bari 1974). Models of the auditory system should attempt to produce a mechanism for the role of the efferent fibers in refining frequency discrimination and identification of signal in noise.

AL Bishop & OW Henson, Jr. 1988. The efferent auditory system in doppler-shift compensating bats. In: PE Nachtigall & PWB Moore, Animal Sonar: Processes and Performance. Plenum Press.

Iurato, S & MD Bari. 1974. Efferent innervation of the cochlea. In: WD Keidel & WD Neff (eds.), Handbook of Sensory Physiology (Vol. 1). Springer-Verlag, pp. 261-282.


inward direction of neural impulses, toward the central nervous system
basilar membrane
a membrane in the cochlea or inner ear dividing the cochlea and supporting the auditory sensory structure
on the other side
toward the top
outward direction of neural impulses, from the central nervous system toward the periphery
supply nerves to
on the same side
organized in ordered layers
a spatial arrangement that provides a “mapping” function
toward the bottom

Wesley R. Elsberry

Falconer. Interdisciplinary researcher: biology and computer science. Data scientist in real estate and econometrics. Blogger. Speaker. Photographer. Husband. Christian. Activist.

2 thoughts on “Wired for Sound: A Q&A Concerning Dolphin Brains

  • 2005/02/26 at 3:05 pm

    TY DR Elsberry, I linked it.

  • 2005/06/23 at 2:54 pm

    Dear Dr. Elsberry:

    I am just beginning to study about dolphins and am interested in analyzing clicks to connect them with images or dolphin “words.” What is the purpose of analyzing dolphin neural network of hearing? I thought that when I did take time to research this, I would analyze the clicks in terms of language. I imagine, though, that knowledge of dolphin neural paths might be useful. As you might have gathered, my ignorance of this is profound. Any comments?

    Kathy Sullivan

Comments are closed.