No published information. In Nicaragua, a juvenile and adult male foraged together (within 50 cm of each other), and performed a "duet" (SW pers. obs.). Unfortunately it is not clear which individual sang which songs in this recording (xeno-canto XC42520).
Limited vocal array consisting of 4 main vocalizations: song, "rah," "didit," and "pew."
Detailed descriptions of the vocal repertoire of the Chestnut-backed Antbird are found in Skutch (1969) and Willis and Oniki (1972), from which much of what follows is drawn.
Song [xeno-canto XC42521, XC3247, Macaulay Library 76682, 28588]: "Loudsong" of Zimmer and Isler (2003). "Two or three full, mellow whistles," transcribed as "fee, few!" or "fee, feh, few!" (Slud 1964, Skutch 1969, Willis and Oniki 1972). The last note is usually slightly down-slurred and may be lower in pitch (Figure 2). Skutch and Willis and Oniki both refer to F. M. Chapman’s characterization of the song as "come here" or "come right here," but we find that "drink beer!" or "drink more beer!" both capture the song’s spirit more closely, and are also more likely to be remembered by students, birders and field assistants struggling to learn many rainforest bird sounds quickly. The first note is more emphatic, with a deliberate, but short (1 s) pause before the next note, which may or may not be of lower pitch. In the 3-note song it is the first syllable that is repeated (the second note sometimes at a higher pitch) and never the last syllable.
"rah": [xeno-canto XC15528, XC30249; Macaulay Library 37762; Figure 3] Skutch described this as "waaa" or "aaaa;" Willis and Oniki referred to this as "rasping." To those familiar with North American birds, this call is not unlike the call of eastern populations of House Wren (Troglodytes aedon), or somewhat similar to the typical call of Gray Catbird (Dumetella carolinensis).
"didit": [Macaulay Library 7209] Skutch (1969) described this as the "alarm note or scold," transcribed as "wittit wittit wittit wittit." Willis and Oniki called this "chipping," and transcribed it as "quit-it." Willis and Oniki describe another vocalization as "rattle" (transcribed as "di’i’i’i’i’i’"). Our impression is that the "rattle" of Willis and Oniki is an extended form of "didit," as perhaps suggested by Skutch’s repetition in his transcription.
"pew": Skutch does not mention this call; Willis and Oniki called this "chirp" (transcribed as "cheup"). These are soft, muted calls that could also be transcribed as "poo," "few," or "tyew," sometimes strung together in warbling ways. Some of these calls are faintly audible in this recording (xeno-canto XC42520).
Willis and Oniki (1972) further describe several other vocalizations that are not frequently heard.
None described. See ‘Species Recognition’ below. Descriptions by Skutch (1969; subspecies occidentalis) and Willis and Oniki (1972; subspecies exsul) agree well. Recordings on commercially available CDs (e.g., Ross and Whitney 1995, Isler and Whitney 2002, Jahn et al. 2002), as well as those archived on xeno-canto, suggest no major variation among subspecies. In Ecuador (subspecies maculifer), may give 3-note song [e.g., Macaulay Library 28482] more frequently than 2-note song (RST, MLB, MJM, pers. obs.).
Skutch (1969) describes decreased vocal activity during the dry season. On Barro Colorado Island (BCI), Panama, Willis and Oniki (1972) note that song activity decreases during the dry, nonbreeding season. Males respond aggressively to playback year round, but females in northern Costa Rica (subspecies exsul) more responsive towards beginning of main breeding season (January – April; SW pers. obs.).
Most vocal at dawn (e.g., Blake 1992), but can be unpredictably heard singing anytime during daylight hours. Willis and Oniki (1972) note "Dark periods after rain and in the early mornings are especially songful."
Places of vocalizing
Little published information; see Willis and Oniki (1972). Songs given in response to playback are typically delivered from a perch (typically < 2m), and not from the ground. Unsolicited songs, however, are not infrequently given while foraging on the ground (SW pers. obs.).
Repertoire and delivery of songs
Limited song repertoire: most often a two-note song, but three and four note songs occasionally heard, where the extra syllables are the first one, repeated. Average delivery rate is 7.6 songs per minute (Willis and Oniki 1972); a maximum song rate of 93 songs hr-1 was reported by Stutchbury et al. (2005) for a male in a territorial dispute, but they did not report the frequency of songs per minute in singing bouts.
Social context and presumed function
Both males and females share the vocal repertoire.
Song: "Loudsong" of Zimmer and Isler (2003). Males sing most; often answered more softly with same song by female. In Costa Rica, a recently divorced female sang loudly and persistently in a manner indistinguishable from male song (SW pers. obs.), but Stutchbury et al. (2005) reported no such behavior for a widowed female they observed. Male song presumably to attract females, defend territories, and also to maintain within-pair contact when out of visual range of each other. Singing can be "contagious:" the sudden song of a Chestnut-backed Antbird in an otherwise silent forest often elicits song from one or more neighbors, sometimes at distances of 300 m or more (SW pers. obs.). Territorial interactions involve a complex and noisy combination of song, rah and didit calls. Response by male to playback within his territory typically swift. In response to playback, males typically do not attack the speaker, as many North American species on the breeding grounds do, although both males and females have been seen on occasion to physically attack the speaker while making prolonged didit vocalizations. A typical response to playback of male is (1) 1-2 brief songs, (2) silent approach of the speaker to within several meters, and (3) constant singing and flying back and forth in the vicinity of (or around) the speaker, looking for the intruder. Females typically respond less readily or aggressively, and if they approach frequently approach silently (omitting step 1 above). Females with an active nest or attendant young seemingly do not respond to playback of any kind (SW pers. obs.).
"rah": Often heard in aggressive interactions with conspecifics, or when the bird is agitated (as perhaps when a person walks by; Willis and Oniki 1972), but often the context is unknown; pairs may also use this as a contact call.
"didit": Given in seemingly stressful interactions (when it may be strung together in many repetitions), such as territorial disputes/interactions. This call is almost invariably given (singly or doubly) when a bird flies across a trail or stream more than 2 or 3 m wide, and is also frequently given upon release after capture in a mist net (SW pers. obs.).
"pew": Willis and Oniki (1972) describe the typical context of these vocalizations: pairs, or adults with attendant young foraging near each other often keep a more or less constant (but rather quiet) chatter going between them using variations on this call, and we have also observed this in Costa Rica (SW).
At the La Selva Biological Reserve in Costa Rica (subspecies exsul), broadcast recordings ("playback") from subspecies exsul, occidentalis, and cassini all elicit territorial responses. The species does not respond to playback of Formicarius analis or Henicorhina leucosticta (SW pers. obs.).