Cochlear Implants and Listening Effort: the Interaction of Cognitive and Sensory Constraints

This study examines how cochlear implant users understand and comprehend speech in realistic communication situations. Through six experiments measuring listening effort via pupillometry and discourse comprehension, we will investigate how linguistic context, cognitive demands, and processing time affect speech understanding in CI users, and in normal-hearing controls) to identify factors underlying communication resilience versus vulnerability and develop improved, ecologically valid assessment and rehabilitation strategies.

Study Overview

• Status: Recruiting
• Conditions: Cochlear Implant Users
• Intervention / Treatment: Behavioral: Experiment 1: Syntactic and Semantic Context; Behavioral: Experiment 2: False Hearing and Context Overuse; Behavioral: Experiment 3: Two-Sentence Problem; Behavioral: Experiment 4: Cascading Effects on Discourse Comprehension; Behavioral: Experiment 5: Self-Paced Discourse Comprehension; Behavioral: Experiment 6: Clinical Application

Detailed Description

This research study examines how adults using cochlear implants (CIs) understand and comprehend speech in realistic communication situations. While current clinical tests focus on how well CI users can recognize single words and simple sentences, this study investigates whether success in recognizing speech sounds actually translates to understanding the meaning and content of longer conversations and discourse.

The study will include six interconnected experiments examining: (1) how CI users use linguistic context (both adaptively and maladaptively) to understand degraded speech; (2) how listening effort affects comprehension when faced with communicative challenges like remembering multiple sentences or narratives; (3) whether giving listeners control over the speed of speech presentation improves comprehension; and (4) a clinically-applicable version of these assessments.

Throughout the experiments, researchers will measure listening effort using pupillometry (tracking pupil dilation as an index of cognitive effort) combined with behavioral measures of speech recognition and comprehension. Comprehension will be assessed using a validated framework that distinguishes between understanding main ideas versus minor details in discourse passages.

The study will include cochlear implant users and normal-hearing adults listening to degraded speech simulations via vocoders. Participants will range in age from 18 to 80 years. A comprehensive baseline battery will assess perceptual abilities (speech recognition, spectral resolution, temporal processing) and cognitive abilities (working memory, processing speed, executive function).

OBJECTIVES:

Primary objectives are to identify mechanisms underlying successful speech comprehension in CI users and to determine factors associated with resilience versus vulnerability to communicative challenges. Secondary objectives include examining relationships between cognitive abilities, listening effort, and discourse comprehension outcomes.

OUTCOMES:

This research is expected to provide more ecologically valid assessment methods for CI users, identify which individuals may benefit from specific communication strategies (such as self-paced speech), and inform development of improved rehabilitation approaches that enhance real-world communication success rather than just word recognition ability.

• Study Type: Interventional
• Enrollment (Estimated): 460
• Phase: Not Applicable

Contacts and Locations

• Study Contact: Name: Mario A. Svirsky, PhD; Phone Number: 212-263-7217; Email: Mario.svirsky@nyulangone.org
• Study Contact Backup: Name: Nicole Capach; Phone Number: 646-501-6905; Email: nicole.capach@nyulangone.org

Study Locations

• United States
  • Massachusetts
    • Waltham, Massachusetts, United States, 02453
      • Recruiting
      • Brandeis University
      • Principal Investigator: Arthur Wingfield
  • New York
    • New York, New York, United States, 10016
      • Recruiting
      • NYU Langone Health
      • Principal Investigator: Mario A. Svirsky, PhD

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: Yes

Description

Inclusion Criteria:

• Subjects will be otherwise healthy normal-hearing and cochlear implant (CI) adult listeners (between 18 and 80 years old).

Exclusion Criteria:

• Individuals below 18 years of age.
• Individuals with evidence of neurologic, vascular or psychiatric disease or dementia, and taking medications that might interfere with task performance.
• Individuals with a history of language disorders (besides those associated with hearing loss for the CI users). Individuals who are non-native speakers of American English.

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Other
• Allocation: Non-Randomized
• Interventional Model: Parallel Assignment
• Masking: None (Open Label)

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: Cochlear Implant Users; Postlingually deaf adults (age 18-80) with at least one year of CI experience. Participants will complete behavioral speech perception and comprehension tasks with pupillometry measurement.	Behavioral: Experiment 1: Syntactic and Semantic Context; Recall of meaningful sentences, anomalous word strings, and unstructured word lists; Measurement of syntactic and semantic gain; Pupillometry during auditory and visual presentation; Behavioral: Experiment 2: False Hearing and Context Overuse; Two-choice word recognition task with semantic priming/luring in multi-talker babble; Three Signal-to-Noise Ratio (SNR) levels (heavy, medium, light noise); Confidence ratings for responses; Pupillometry measurement; Behavioral: Experiment 3: Two-Sentence Problem; Speech recognition and recall of single sentences vs. paired sentences; Manipulation of inter-sentence semantic predictability (high vs. low); Four test conditions: 1-sentence, 2-sentences, 2-sentences+pre-prompt, 2-sentences+post-prompt; Pupillometry during task; Behavioral: Experiment 4: Cascading Effects on Discourse Comprehension; Recall of 27 narrative passages (67-97 words each); Propositional analysis scoring (main ideas, mid-level ideas, details); Measurement of semantic hierarchy effect; Pupillometry during listening; Behavioral: Experiment 5: Self-Paced Discourse Comprehension; 24 discourse passages (150 words each): 12 narrative, 12 expository; Continuous presentation vs. self-paced presentation (stops at clause/sentence boundaries); Measurement of pause times and comprehension recall; Pupillometry during task; Behavioral: Experiment 6: Clinical Application; Self-Paced Sentence Comprehension; Sentences with varying syntactic complexity (active-conjoined, subject-relative, object-relative); Continuous vs. self-paced (with pause at major clause boundary) presentation; True/false comprehension verification statements; Pupillometry measurement
Active Comparator: Normal-Hearing Controls (Vocoder Simulation); Normal-hearing adults (age 18-80) listening to degraded speech via 4- and 8-channel vocoders. Participants will complete the same behavioral tasks as CI users but with (or without) acoustically degraded speech simulation.	Behavioral: Experiment 1: Syntactic and Semantic Context; Recall of meaningful sentences, anomalous word strings, and unstructured word lists; Measurement of syntactic and semantic gain; Pupillometry during auditory and visual presentation; Behavioral: Experiment 2: False Hearing and Context Overuse; Two-choice word recognition task with semantic priming/luring in multi-talker babble; Three Signal-to-Noise Ratio (SNR) levels (heavy, medium, light noise); Confidence ratings for responses; Pupillometry measurement; Behavioral: Experiment 3: Two-Sentence Problem; Speech recognition and recall of single sentences vs. paired sentences; Manipulation of inter-sentence semantic predictability (high vs. low); Four test conditions: 1-sentence, 2-sentences, 2-sentences+pre-prompt, 2-sentences+post-prompt; Pupillometry during task; Behavioral: Experiment 4: Cascading Effects on Discourse Comprehension; Recall of 27 narrative passages (67-97 words each); Propositional analysis scoring (main ideas, mid-level ideas, details); Measurement of semantic hierarchy effect; Pupillometry during listening; Behavioral: Experiment 5: Self-Paced Discourse Comprehension; 24 discourse passages (150 words each): 12 narrative, 12 expository; Continuous presentation vs. self-paced presentation (stops at clause/sentence boundaries); Measurement of pause times and comprehension recall; Pupillometry during task; Behavioral: Experiment 6: Clinical Application; Self-Paced Sentence Comprehension; Sentences with varying syntactic complexity (active-conjoined, subject-relative, object-relative); Continuous vs. self-paced (with pause at major clause boundary) presentation; True/false comprehension verification statements; Pupillometry measurement

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Percent correct Consonant-Nucleus-Consonant (CNC) words across experiments 1-3	Participants' ability to recognize individual spoken words will be assessed using the Consonant-Nucleus-Consonant (CNC) word test. Outcome is based on the percent of key words correctly repeated. Higher score indicate better speech understanding.	End experiments 1-3 (up to 9 hours)
Percent correct AzBio sentences across experiments 1-3	The AzBio sentence test consists of 15 lists of 20 sentences each. AzBio sentences are spoken by different talkers in a conversational style with limited contextual cues that the listener can use to predict or 'fill in' unintelligible words. Outcome is based on the percent of AzBio sentences correctly repeated. Higher score indicate better speech understanding.	End experiments 1-3 (up to 9 hours)
Error rates for word identification tasks across experiments 1-3		End experiments 1-3 (up to 9 hours)
Percentage of propositions recalled from narrative passages	Subjects will hear 24 150- word passages, with one half of the passages being narrative (tell a story with a setting, theme, plot, etc.) and the other half being expository (e.g. instructions from a health care provider). After each passage the subject will be asked to recall as much of the passage content as possible. Higher score indicate better speech understanding.	End of experiment 4 (up to 3 hours)
Ratio of main idea recall to detail recall	Subjects will hear recorded narratives and will be asked to recall them as best as they can. The narratives will include main ideas (defined as propositions whose arguments are directly related to the overall meaning of the passage), mid-level propositions (those that take main propositions as their arguments), and details (propositions that take mid-level or other minor propositions as their arguments. Ratio of main idea recall to detail recall will be assessed.	End of experiment 4 (up to 3 hours)
Percentage of true/false statements identified	Subjects will hear sentences with syntactic forms known to be graded in comprehension difficulty: active-conjoined sentences (The author insulted the critic // and the critic hired a lawyer), subject-relative sentences (The author that insulted the critic // hired a lawyer), and object-relative sentences (The author that the critic insulted // hired a lawyer). The subject will hear an equal number of each sentence type, where half are presented with self-pacing, and half are presented without interruption. After each sentence the subject will be presented that they must indicate is true or false about the sentence they just heard.	End of experiment 6 (up to 3 hours)
Percent correct increase from unstructured word lists to anomalous sentences (syntactic gain)	Participants will hear 30 meaningful, eight-word sentences to begin. The words from this list of sentences will then be used to create the unstructured word lists and syntactic strings, such that each word would be heard equally as often across the three stimulus conditions. Therefore, each subject will hear a total of 30 meaningful sentences, 30 anomalous strings, and 30 unstructured lists, presented in a counter-balanced design. Recall accuracy for the three conditions will be compared.	End of experiment 1 (up to 3 hours)
Percent correct increase from anomalous sentences to meaningful sentences (semantic gain).	Participants will hear 30 meaningful, eight-word sentences to begin. The words from this list of sentences will then be used to create the unstructured word lists and syntactic strings, such that each word would be heard equally as often across the three stimulus conditions. Therefore, each subject will hear a total of 30 meaningful sentences, 30 anomalous strings, and 30 unstructured lists, presented in a counter-balanced design. Recall accuracy for the three conditions will be compared.	End of experiment 1 (up to 3 hours)
Number semantically-driven misrecognitions in lure conditions	Subjects will hear word pairs, with the second (target) word of the pair presented in multi-talker babble. There will be three conditions: Neutral prime (the first word is unrelated to the target word, e.g. Jaw-PASS), Semantic prime (the two words are semantic associates, e.g. Row - BOAT), and Semantic lure (the target word is a semantic associate of a similar word e.g. Row - GOAT, where GOAT is a lure for BOAT), thus putting the semantic context in conflict with successful perception. Number semantically-driven misrecognitions in lure conditions will be assessed.	End of experiment 2 (up to 3 hours)
Frequency of semantically-driven misrecognitions in lure conditions	Subjects will hear word pairs, with the second (target) word of the pair presented in multi-talker babble. There will be three conditions: Neutral prime (the first word is unrelated to the target word, e.g. Jaw-PASS), Semantic prime (the two words are semantic associates, e.g. Row - BOAT), and Semantic lure (the target word is a semantic associate of a similar word e.g. Row - GOAT, where GOAT is a lure for BOAT), thus putting the semantic context in conflict with successful perception. Frequency of semantically-driven misrecognitions in lure conditions will be assessed.	End of experiment 2 (up to 3 hours)
Difference in recall accuracy between single-sentence and two-sentence conditions	Subjects will undergo four test conditions (two will require remembering only one sentence and the other two will require remembering two sentences). For each condition there will be two combinations of semantic association between the first and second sentence (half will have high predictability, and the other half will have low predictability). Difference in recall accuracy between single-sentence and two-sentence conditions will be assessed.	End of experiment 3 (up to 3 hours)
Difference in comprehension improvement (percentage of proposition's correctly recalled) with self-paced vs. continuous presentation	Subjects will hear 24 150- word passages, with one half of the passages being narrative (tell a story with a setting, theme, plot, etc.) and the other half being expository (e.g. instructions from a health care provider). After each passage the subject will be asked to recall as much of the passage content as possible. Subjects will be presented with equal numbers of "self-paced" passages (stopping at major clauses and sentence boundaries, subject decides when to continue) and passages presented without interruption.	End of experiment 5 (up to 3 hours)

Collaborators and Investigators

• Sponsor: NYU Langone Health
• Collaborators: National Institute on Deafness and Other Communication Disorders (NIDCD)
• Investigators: Principal Investigator: Mario A. Svirsky, PhD, NYU Langone Health

Study record dates

Study Major Dates

• Study Start (Actual): January 2, 2025
• Primary Completion (Estimated): January 2, 2030
• Study Completion (Estimated): January 2, 2030

Study Registration Dates

• First Submitted: December 9, 2025
• First Submitted That Met QC Criteria: December 9, 2025
• First Posted (Actual): December 12, 2025

Study Record Updates

• Last Update Posted (Actual): December 12, 2025
• Last Update Submitted That Met QC Criteria: December 9, 2025
• Last Verified: December 1, 2025

More Information

Terms related to this study

• Keywords: Cochlear implant; CI
• Other Study ID Numbers: 18-01806; 2R01DC016834-06A1 (U.S. NIH Grant/Contract)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Plan Description: The de-identified participant data from the final research dataset will be shared upon reasonable request. Data will be made available as soon as possible or at the time of associated publications. All data to be shared will be shared by the close of the award. Data will be made available, at minimum, for seven years at a third party website (https://databrary.org/)
• IPD Sharing Time Frame: Data will be made available as soon as possible or at the time of associated publications. All data to be shared will be shared by the close of the award. Data will be made available, at minimum, for seven years.
• IPD Sharing Access Criteria: The investigator who proposed to use the data will be granted access upon reasonable request. Data will be available for at least 7 years at a third party website (https://databrary.org/)
• IPD Sharing Supporting Information Type: STUDY_PROTOCOL; SAP

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No