Inner Speech

Hearing Words off the Page

Hearing inner voices while reading speech quotes is a phenomenon closely tied to inner speech, the internal dialogues that engages various cognitive and neural mechanisms. This experience can vary in tone, pitch, or accent based on the reader’s familiarity with characters or context, and it provides valuable insights into the interplay between language and perception. Inner speech during reading can affect comprehension, emotional engagement, and memory retention, and its experience can differ among individuals due to factors like reading proficiency, cognitive style, and mood. For example, some people report a more vivid inner voice when emotionally engaged with the text.

At LISN, our research focuses on the experience of hearing inner voices during silent reading of speech quotes. Empirical evidence from our lab indicates increased neural activity in the auditory cortex (Yao, Belin, & Scheepers, 2011), more synchronous auditory cortical oscillations (Yao et al., 2021), and eye movements that align with the speed of the inner voice (Yao & Scheepers, 2011). We explore underlying mechanisms and contributing factors, such as the hypothesis that learning to speak and read aloud may establish an automatic link between text and auditory experiences. Another line of inquiry considers that this inner voice simulates social interactions (Alderson-Day et al., 2020), potentially reinforced by the brain’s reward system as a prosocial mechanism. Through these research avenues, LISN aims to deepen our understanding of this complex aspect of language processing and cognition.

Varieties of Inner Speech

Inner speech manifests in various subtypes, each with distinct characteristics. Dialogic inner speech involves an internal dialogue where different perspectives are represented. Condensed inner speech is more abbreviated and lacks the full syntactic structure of external speech. Evaluative inner speech involves self-assessment or self-criticism, and motivational inner speech serves to encourage or motivate the individual. These subtypes offer valuable insights into the complexities of human cognition. To study them empirically, researchers use a range of methods. Self-report questionnaires capture individual experiences but can be subject to self-report biases. Experience sampling involves prompting participants at random times to report their inner speech, aiming to capture real-time data. Neuroimaging techniques like fMRI and EEG identify the brain regions activated during different types of inner speech. Think-aloud protocols involve participants verbalising their thought processes during tasks, capturing the structure and content of inner speech.

At LISN, our research employs multiple methods to delve into inner speech subtypes. We use experience sampling in a range of cognitive tasks to understand the prevalence of inner speech subtypes in different situations. Think-aloud protocols help us explore the phenomenology of inner speech across various task conditions. Additionally, we use machine learning and EEG to identify the neural underpinnings of different inner speech subtypes. Through these research avenues, LISN aims to offer a comprehensive understanding of this complex aspect of language processing and cognition.

Neural Mechanisms of Inner Speech

Inner speech, the internal dialogue or monologue that individuals engage in, is a complex neural activity with various subtypes and functions. The neural mechanisms behind inner speech involve multiple brain regions, including the left inferior frontal gyrus (IFG) for speech production and the superior temporal gyri/sulci (STG/STS) for speech comprehension. These regions often interact with other areas responsible for memory, attention, and emotional processing, such as the prefrontal cortex and the amygdala. Neuroimaging techniques like functional magnetic resonance imaging (fMRI) and electroencephalography (EEG) have been instrumental in shedding light on these neural correlates. For example, fMRI studies have shown increased activity in the left IFG and STG during tasks requiring inner speech, while EEG studies have captured the real-time neural activity associated with different types of inner dialogue.

At LISN, our research focuses on understanding these neural mechanisms and how they support inner speech’s diverse phenomenology (Pratts, Pobric, & Yao, 2023). We’ve found increased neural activity in auditory regions through fMRI (Yao, Belin, & Scheepers, 2011) and more synchronous auditory cortical oscillations in EEG (Yao et al., 2021) associated with inner speech during quote reading. Our meta-analysis suggests that inner speech may be supported by both a motor-driven pathway, involving the left IFG, and an auditory pathway, involving auditory perceptual areas. These pathways could potentially underpin different inner speech subtypes (Pratts, Pobric, & Yao, 2023). We are also employing machine learning to identify the neural features that characterise these subtypes, aiming for a more nuanced understanding of inner speech and its role in cognition. Through these research avenues, LISN seeks to deepen our understanding of the neural mechanisms of inner speech and its various subtypes.

Inner Speech in Self-Consciousness

Inner speech serves as a critical interface between thought and consciousness, mediating self-awareness, intentionality, and subjective experience. While it has been a subject of philosophical inquiry for centuries, empirical research is relatively new. Studies have explored how inner speech contributes to different states and levels of consciousness, such as focused attention and mind-wandering. Understanding its role in consciousness can offer insights into self-awareness, intentional action, and even the boundaries of conscious experience. It also has implications for mental health, as disruptions in inner speech often occur in conditions like schizophrenia and certain types of depression.

At LISN, we aim to empirically test the causal role of inner speech in self-awareness, capitalising on methodological advances and the discovery of aphantasia, a condition characterised by an inability to create visual imagery. We plan to objectively measure inner speech abilities in a range of tasks and use EEG to monitor inner speech in self-processing. Our research will pioneer data-driven classification of brain states related to inner speech and self-awareness. This approach could offer new prospects for understanding normal and altered states of consciousness, as well as related mental disorders like rumination, depression, and auditory verbal hallucinations. Through these research avenues, LISN seeks to deepen our understanding of the complex relationship between inner speech and consciousness.